Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Genetic and epigenetic profiling of placental insufficiency : identifying biomarkers of preeclampsia… Wilson, Samantha Lea 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2018_february_wilson_samantha.pdf [ 6.32MB ]
Metadata
JSON: 24-1.0361940.json
JSON-LD: 24-1.0361940-ld.json
RDF/XML (Pretty): 24-1.0361940-rdf.xml
RDF/JSON: 24-1.0361940-rdf.json
Turtle: 24-1.0361940-turtle.txt
N-Triples: 24-1.0361940-rdf-ntriples.txt
Original Record: 24-1.0361940-source.json
Full Text
24-1.0361940-fulltext.txt
Citation
24-1.0361940.ris

Full Text

  GENETIC AND EPIGENETIC PROFILING OF PLACENTAL INSUFFICIENCY: IDENTIFYING BIOMARKERS OF PREECLAMPSIA AND INTRAUTERINE GROWTH RESTRICTION by  Samantha Lea Wilson  B.Sc., The University of Western Ontario, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Medical Genetics)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   December 2017  © Samantha Lea Wilson, 2017 ii  Abstract  Preeclampsia (PE), characterized by maternal hypertension and proteinuria, is the leading cause of maternal and perinatal morbidity and mortality worldwide. PE can be subdivided into early-onset PE (EOPE), diagnosis <34 weeks, and late-onset PE (LOPE), diagnosis ≤34 weeks. Intrauterine growth restriction (IUGR), pathologically poor fetal growth, often co-occurs with PE. Both conditions are thought to be due to the placenta, the organ responsible for oxygen and nutrient transport to the fetus, not functioning adequately; this is referred to as placental insufficiency (PI). Currently there is no consistent clinical test to predict pregnancies at risk of these conditions.   This dissertation investigated placental genetic and epigenetic profiles to assess their utility to identify novel protein biomarkers in maternal blood and to subclassify PE and IUGR placentas. DNA methylation (DNAm) in the placenta at delivery correlated to second trimester Inhibin alpha and third trimester fibronectin  levels in maternal blood, indicating that DNAm alterations may be useful to identify novel biomarkers. Widespread DNAm alterations were identified in EOPE placentas, many of which validated in an independent cohort. These changes can be used to refine clinical diagnoses. Placental DNAm signatures can help to refine clinical diagnoses to create more homogenous pathological groups, aiding in more robust predictive algorithms for PI conditions.   Telomere length (TL) was also assessed as a potential biomarker, as it has been reported altered in placentas of PE and IUGR pregnancies. This is thought to be due to the hypoxic environment iii  associated with PI which results in DNA damage, and shortened TL. However, correcting for placental gestational age and fetal sex, TL was not shorter in LOPE or IUGR in our population. EOPE placentas trended towards shorter TL (p=0.1). Overall, these studies found that TL is not likely to be an informative marker of PI. These studies show the potential utility of DNAm in identifying novel protein biomarkers in maternal blood, refining clinical diagnoses to create more homogenous subtypes of PI, or as a potential biomarker itself for identifying pregnancies at risk of PI conditions.     iv  Lay Summary Health complications during pregnancy affect both the mother and baby. In ~10% of pregnancies babies fail to grow as expected. Slow growth can be seen with high blood pressure and high protein in the mother’s urine; called preeclampsia. These complications occur when the placenta, the organ that is responsible for regulating nutrients and oxygen transport to the baby, functions inadequately. Both of these complications can negatively affect the child’s health. Currently there is no conclusive method to predict which women will develop these complications. This dissertation investigates potential biomarkers to predict pregnancies at risk of these conditions. I found that DNA methylation, modifications to DNA that change how the gene operates, can be used to identify new biomarkers in the mother’s blood and help properly diagnose these conditions. Telomere length, repetitive regions at the end of chromosomes are likely not sufficient indicators of placenta not functioning properly.  v  Preface Parts of this dissertation were previously published as follows: 1.! A version of the Introduction and Discussion has been submitted as an invited review for publication by SL Wilson and WP Robinson (2017).  2.! A version of Chapter 3 has been published in BMC Medical Genetics. Wilson SL, Blair JD, Hogg K, Langlois S, von Dadelszen P, Robinson WP. Placental DNA methylation at term reflects maternal serum levels of INHA and FN1, but not PAPPA, early in pregnancy. BMC medical genetics. 2015 Dec 11;16(1):111. •! © Wilson et al., under the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/). Data, text, figures and tables published in this article are contained in Chapter 3. 3.! A version of Chapter 4 has been published in Human Molecular Genetics. Wilson, SL, Leavey, K., Cox, BJ., Robinson, WP. Mining DNA methylation alterations towards a classification of placental pathologies. Human Molecular Genetics. Oct 2017. •! © Wilson et al., under the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/). Data, text, figures and tables published in this article are contained in Chapter 4.  A pre-print of this publication is available on bioarxiv, copyright license is available under a CC-BY-ND 4.0 International license.  4.! A version of Chapter 5 has been published in Placenta. Wilson SL, Liu Y, Robinson WP. Placental telomere length decline with gestational age differs by sex and TERT, DNMT1, and DNMT3A DNA methylation. Placenta. 2016 Dec 31;48:26-33. vi  •! © 2016 Elsevier Ltd., under the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/). Data, text, figures and tables published in this article are contained in Chapter 5. As Chapter 3, 4, and 5 have remained relatively unchanged from their published versions (Chapters 3-5). I have preserved the use of plural first person nouns. In the remainder of the dissertation, I use first person pronouns.   Parts of this dissertation were performed by collaborators: •! Chapter 3 o! A subset of maternal plasma samples were ascertained by Dr. Peter von Dadelszen, others were recruited by members of the Robinson lab. Women were recruited either by the Robinson lab research coordinator (various) or referring physicians including Dr. Sylvie Langlois, Dr. Hayley Bos, Dr. Peter von Dadelszen. In addition some placental samples were obtained by Dr. Deborah McFadden and the Dept. of Pathology at BC Children’s and Women’s Health Centre. Placental sampling and DNA extraction were completed by Ruby Jiang and Dr. Maria Peñaherrera. All co-authors edited and contributed to the manuscript. Ethics approval was granted by the University of British Columbia and BC Women’s and Children’s Hospital Research Ethics Board (H04-70488). Experimental design and hypothesis were conceived by WP Robinson and myself. K Hogg helped design the pyrosequencing primers for PAPPA. I performed the experiments and data analysis.  •! Chapter 4 vii  o! Recruitment of women and placental processing in the discovery cohort were completed as per Chapter 3. Recruitment of samples from the validation cohort was completed by Katherine Leavey and Brian Cox. Early-onset preeclampsia and preterm controls samples on the Illumina Infinium HumanMethylation450 array were run by John Blair and Dr. Maria Peñaherrera. Late-onset preeclampsia, intrauterine growth restriction, and term control placentas on the Illumina Infinium HumanMethylation450 array were run with the help of Dr. Maria Peñaherrera. Experimental design and hypothesis were conceived by WP Robinson and myself. I performed the data analysis. All co-authors edited and contributed to the manuscript. Ethics approval was granted by the University of British Columbia and BC Women’s and Children’s Hospital Research Ethics Board (H04-70488). •! Chapter 5 o! Recruitment of women and placental processing in the third trimester cohort were completed as per Chapter 3. In addition, samples from earlier in pregnancy were obtained with the help of Dr. Alex Beristain, Dr. Deborah McFadden and the Dept. of Pathology at BC Children’s and Women’s Health Centre. Placental sampling and DNA extraction were completed by Ruby Jiang and Dr. Maria Peñaherrera. Telomere length measurements by qPCR were complete with the help for Yao Liu. Early-onset preeclampsia and preterm controls samples on the Illumina Infinium HumanMethylation450 array were run by John Blair and Dr. Maria Peñaherrera. Late-onset preeclampsia, intrauterine growth restriction, and term control placentas on the Illumina Infinium HumanMethylation450 array viii  were run with the help of Dr. Maria Peñaherrera. Experimental design and hypothesis were conceived by WP Robinson and myself. I performed the data analysis. All co-authors edited and contributed to the manuscript. Ethics approval was granted by the University of British Columbia and BC Women’s and Children’s Hospital Research Ethics Board (H04-70488).    Table of Contents  Abstract .......................................................................................................................................... ii"Lay Summary ............................................................................................................................... iv"Preface .............................................................................................................................................v"Table of Contents ......................................................................................................................... ix"List of Tables ................................................................................................................................xv"List of Figures ............................................................................................................................. xvi"List of Symbols .......................................................................................................................... xvii"List of Abbreviations ............................................................................................................... xviii"Acknowledgements .................................................................................................................... xxi"Chapter 1: Introduction ................................................................................................................1"1.1" Placental insufficiency .................................................................................................... 1"1.1.1" Preeclampsia and other hypertensive disorders of pregnancy .................................... 1"1.1.2" Intrauterine growth restriction .................................................................................... 6"1.2" The placenta and its development ................................................................................. 10"1.2.1" Placental structure ..................................................................................................... 10"1.2.2" Placental development .............................................................................................. 10"1.2.2.1" Implantation .......................................................................................................10"1.2.2.2" Angiogenesis ......................................................................................................13"1.2.2.3" Nutrient transport ...............................................................................................13"1.2.2.4" Paracrine, endocrine, autocrine and immune functions in the placenta ............15"1.2.2.5" Placental development in preeclampsia and intrauterine growth restriction .....15"x  1.3" Current screening for placental insufficiency ............................................................... 16"1.3.1" Imaging tools ............................................................................................................ 16"1.3.1.1" Ultrasound ..........................................................................................................16"1.3.1.2" Magnetic resonance imaging and positron emission topograph ........................17"1.3.2" Maternal serum markers ........................................................................................... 19"1.4" Molecular studies of the placenta in placental insufficiency ........................................ 19"1.4.1" Chromosomal associations ........................................................................................ 19"1.4.2" Genetic variants ........................................................................................................ 20"1.4.3" Telomere length variation ......................................................................................... 22"1.4.4" Transcriptomic alterations ........................................................................................ 22"1.4.4.1" Altered gene expression in preeclampsia and intrauterine growth restriction ...22"1.4.4.2" Altered miRNA profiles in preeclampsia and intrauterine growth restriction ..23"1.5" DNA methylation .......................................................................................................... 24"1.5.1" Influencing variables of DNA methylation ............................................................... 25"1.5.1.1" Sex specific DNA methylation ..........................................................................25"1.5.1.2" Environmental exposures ...................................................................................26"1.5.1.3" Disease state .......................................................................................................26"1.5.1.4" Tissue-specific DNA methylation .....................................................................27"1.5.1.5" Age associated DNA methylation .....................................................................28"1.5.2" The placental methylome .......................................................................................... 29"1.5.3" DNA methylation in preeclampsia and intrauterine growth restriction .................... 30"1.5.4" Measuring DNA methylation .................................................................................... 31"1.5.4.1" Genome-wide DNA methylation measures .......................................................31"xi  1.5.4.2" Candidate gene DNA methylation measures .....................................................32"1.6" Research hypothesis and objectives .............................................................................. 32"Chapter 2: Methods .....................................................................................................................35"2.1" Placental sampling ........................................................................................................ 36"2.2" Clinical definitions of preeclampsia and IUGR ............................................................ 36"2.3" DNA methylation microarray- Illumina Infinium HumanMethylation450 array ......... 37"Chapter 3: Placental DNA methylation at term reflects maternal serum levels of INHA and FN1, but not PAPPA, early in pregnancy. .................................................................................41"3.1" Introduction ................................................................................................................... 41"3.2" Methods......................................................................................................................... 42"3.2.1" Sample information ................................................................................................... 42"3.2.2" Gene expression analysis .......................................................................................... 43"3.2.3" DNA methylation analysis ........................................................................................ 43"3.2.3.1" Illumina Infinium HumanMethylation450 beadchip array ................................43"3.2.3.2" Bisulfite pyrosequencing ...................................................................................43"3.2.4" Candidate DNA methylation selection ..................................................................... 44"3.2.5" Maternal blood protein measurements ...................................................................... 45"3.2.6" Statistical analysis ..................................................................................................... 48"3.3" Results ........................................................................................................................... 48"3.3.1" Candidate site selection and characteristics .............................................................. 48"3.3.2" Is DNA methylation at candidate sites inversely correlated with gene expression? 51"3.3.3" What clinical factors are associated with DNA methylation at candidate sites? ...... 53"3.3.4" What clinical factors are associated with protein concentration in maternal blood? 54"xii  3.3.5" What is the relationship between DNAm and maternal serum levels? ..................... 55"3.3.6" Are there any differences in protein levels between case and control placentas? .... 55"3.4" Discussion ..................................................................................................................... 56"Chapter 4: Utility and validation in placental DNA methylation profiling of preeclampsia and intrauterine growth restriction ...........................................................................................61"4.1" Introduction ................................................................................................................... 61"4.2" Methods......................................................................................................................... 62"4.2.1" Sample information ................................................................................................... 62"4.2.1.1" Discovery cohort ................................................................................................62"4.2.1.2" Validation cohort ...............................................................................................65"4.2.1.3" DNA methylation analysis .................................................................................65"4.2.1.4" Differential DNA methylation analysis .............................................................66"4.2.1.5" Clustering analysis .............................................................................................67"4.3" Results ........................................................................................................................... 68"4.3.1" Widespread DNAm changes are associated with EOPE but not LOPE and nIUGR in our Discovery cohort ............................................................................................................. 68"4.3.2" Validation of the EOPE hits in an independent cohort ............................................. 70"4.3.3" Effects of varying validation criteria ........................................................................ 73"4.3.4" Hierarchical Clustering ............................................................................................. 73"4.3.5" Cluster gene ontology ............................................................................................... 77"4.4" Discussion ..................................................................................................................... 78"Chapter 5: Placental telomere length decline with gestational age differs by sex and TERT, DNMT1, and DNMT3A DNA methylation .................................................................................83"xiii  5.1" Introduction ................................................................................................................... 83"5.2" Methods......................................................................................................................... 85"5.2.1" Sample Information .................................................................................................. 85"5.2.2" Telomere length measures ........................................................................................ 86"5.2.3" DNA methylation analysis ........................................................................................ 86"5.2.4" Statistical analysis ..................................................................................................... 87"5.3" Results ........................................................................................................................... 87"5.3.1" Telomere length is associated with gestational age and fetal sex ............................. 87"5.3.2" Telomere length does not differ between EOPE, LOPE, IUGR and controls .......... 90"5.3.3" Telomere length is associated with DNA methylation at TERT, DNMT1, and DNMT3A ............................................................................................................................... 92"5.3.4" Genome-wide DNA methylation analysis ................................................................ 96"5.4" Discussion ..................................................................................................................... 96"Chapter 6: Discussion ..................................................................................................................99"6.1" Summary of Dissertation .............................................................................................. 99"6.2" Significance and contribution ....................................................................................... 99"6.3" Strengths and limitations ............................................................................................. 101"6.4" Approaches and considerations in reproducibility of DNA methylation studies ....... 103"6.5" Understanding limitations and responsible reporting of results in EWAS studies ..... 105"6.6" Future directions of preeclampsia and intrauterine growth restriction prediction ...... 108"Bibliography ...............................................................................................................................110"Appendices ..................................................................................................................................166"Appendix A Supplementary tables and figures for Chapter 1 ................................................ 166"xiv  Appendix B Supplementary tables and figures for Chapter 2 ................................................ 192"Appendix C Supplementary tables and figures for Chapter 3 ................................................ 193"Appendix D Supplementary figures and tables for Chapter 4 ................................................ 202" xv  List of Tables Table 1.1. Clinical diagnosis criteria of preeclampsia worldwide. ................................................. 5"Table 1.2. Methods of transport utilized by the placenta. ............................................................. 14"Table 3.1. Candidate CpG sites chosen from Blair et al. (2013) for follow-up ............................ 44"Table 3.2. Samples used for pyrosequencing and measuring maternal FN1 protein levels. ........ 46"Table 3.3. Univariate linear analysis results for DNA methylation vs. clinical parameters. ........ 53"Table 3.4. Univariate linear analysis results for protein Levels vs. clinical parameters. .............. 54"Table 4.1. Sample information on the discovery and validation cohorts. ..................................... 63"Table 4.2. Clinical information on samples assigned to methylation clusters. ............................. 76"Table 5.1. Placental samples and clinical data for which telomere length was measured. ........... 85" xvi  List of Figures Figure 1.1 Risk factors for preeclampsia (PE) and intrauterine growth restriction (IUGR). .......... 9"Figure 1.2. Cell differentiation of cell types that comprise fetal tissues and the placenta. ........... 12"Figure 3.1. DNA methylation measures at candidate sites. .......................................................... 50"Figure 3.2.  Correlation between DNA methylation and left) gene expression, right) protein levels. ............................................................................................................................................ 52"Figure 3.3. Fibronectin 1 (FN1) levels in each pathological group. ............................................. 56"Figure 3.4. Processes that may influence the relationship between DNA methylation, gene expression and protein expression. ............................................................................................... 58"Figure 4.1. Volcano plots depicting differentially methylated sites. ............................................ 69"Figure 4.2. Validation of DNA methylation alterations. .............................................................. 72"Figure 4.3. Hierarchical clustering (Euclidean) on the 599 validated hits in both the discovery (left) and validation (right) cohorts. .............................................................................................. 75"Figure 5.1. Fetal sex is associated with telomere length. .............................................................. 88"Figure 5.2. Variables assessed as potential confounders with telomere length. ........................... 89"Figure 5.3. Boxplots depicting telomere length (T/S ratio) between pathological groups. .......... 91"Figure 5.4. Sites in TERT, DNMT1 and DNMT3A with association between DNA methylation and telomere length.. ..................................................................................................................... 93"Figure 5.5. DNA differences within TERT, DNMT1, and DNMT3A. ........................................... 95" xvii  List of Symbols β  DNA methylation measure Δβ  Change in DNA methylation xviii  List of Abbreviations 450K  Illumina Infinium HumanMethylation450 array AC  Abdominal circumference AFP  Alpha-fetoprotein BMI  Body mass index  BP  Blood pressure CG  Cytidine-guanosine CoV  Coefficient of the variance CpG  Cytidine-phosphate-guanosine CPM  Confined placental mosaicism DNA  Deoxyribose nucleic acid DNAm DNA methylation DNMT DNA methyltransferases ENCODE Encyclopedia of gene elements EOPE  Early-onset preeclampsia EPIC  Illumina Infinium MethylationEPIC array EVT  Extravillous trophoblast  EWAS  Epigenome-wide association studies FDR  False discovery rate FN1  Fibronectin 1 GWAS Genome-wide association studies  hCG  Human chorionic gonadotropin HELLP Hemolysis, elevated liver enzymes, and low platelet count xix  HLA  Human leukocyte antigen ICM  Inner cell mass INHA  Inhibin A INHBA Inhibin beta alpha IUGR  Intrauterine growth restriction LOPE  Late-onset preeclampsia miRNA microRNA MLPA  Multiplex ligation dependent probe amplification MoM  Multiple of the median nIUGR Normotensive IUGR PAPPA Pregnancy associated plasma protein A PCR  Polymerase chain reaction PE  Preeclampsia PIH  Pregnancy induced hypertension PLGF  Placental growth factor PMD  Partially methylated domain qPCR  Quantitative polymerase chain reaction RNA   Ribonucleic acid SD  Standard deviation SGA  Small for gestational age SNP  Single nucleotide polymorphism SOGC  The Society of Obstetricians and Gynecologists of Canada SPH  Symphysis fundal height xx  SWAN Subset within-array normalization TERC  Telomerase RNA component TERT  Telomerase reverse transcriptase TL  Telomere length WHO  World Health Organization wks  Weeks xxi  Acknowledgements I would first like to acknowledge my graduate supervisor, Dr. Wendy Robinson, for her patience, mentorship, and guidance to allow me to develop and grow as both a scientific researcher and as an individual. Thank you to both past and present Robinson lab members (Maria, Ruby, Kristal, Johanna, Kirsten, Courtney, John, Magda, Olivia, Irina, Chaini, Giulia, Victor, Desmond, Yao, and Li Qing), who have helped along the way both with experiments and challenging my perspective of science, this thesis could not have been done without you. I would also like to acknowledge my advisory committee, Dr. Sylvie Langlois, Dr. Julian Christians, Dr. Louis Lefebvre, and Dr. Sara Mostafavi. Thank you for your continued support and guidance throughout my PhD career. A special thanks to Rachel Edgar, for the many science discussions, sharing of code, support, and glasses of wine; and to Jack Hickmott for the immense support (much of which was getting up and down very steep mountains), encouragement, and all the dinners while we drank the wine. Thank you to my family: Mom (Dave), Dad, Katie Bird, Grandma, Tampa, Page, Marita and my many aunts, uncles and cousins, for all your love, reassurance and strength. Finally, I would like to acknowledge my 9th grade science teacher, Mr. Robert Munro, who encouraged me to continue my science education and think of science as a career.  Thank you to the institutions that provided financial support for the work covered in this dissertation: BC Children’s Hospital Research Institute (Graduate Studentship), University of British Columbia (4 Year Doctoral Fellowship), and CIHR operating grant to support research costs. 1  Chapter 1:!Introduction  1.1! Placental insufficiency The term ‘placental insufficiency’ encompasses many conditions in which the placenta does not function adequately. Understanding pathogenesis and screening for placental insufficiency conditions is important as pregnancies with these conditions are at an increased risk of preterm birth which is associated with both short and long term adverse health effects on the baby. Most commonly reported are hypertensive disorders of pregnancy and fetal growth restriction, specifically intrauterine growth restriction (IUGR). Hypertensive disorders of pregnancy can be divided into chronic hypertension,pregnancy induced hypertension (PIH), Hemolysis elevated liver enzymes and low platelet count (HELLP) syndrome, and preeclampsia (PE).  1.1.1! Preeclampsia and other hypertensive disorders of pregnancy Pregnancy induced hypertension (PIH), is defined by high blood pressure (BP) first observed in pregnancy, but not accompanied by any other clinical symptoms and returning to normal within 12 weeks of delivery (1). Hemolysis elevated liver enzymes and low platelet count"(HELLP) syndrome commonly co-occurs with PE, but also occurs as non-hypertensive HELLP syndrome (2). It is characterized by Hemolysis, elevated liver enzymes, and low platelet count, giving rise to the name HELLP. HELLP syndrome can cause severe liver damage and may result in severe adverse health outcomes on both the mother and the fetus (2). Preeclampsia (PE) is a multi-system, maternal hypertensive disorder commonly characterized by high BP and proteinuria, after 20 weeks’ gestation (3). PE occurs in 2-8% of pregnancies worldwide, and is the leading cause of maternal and perinatal morbidity and mortality (4). While high BP is consistently 2  considered to be equal to or greater than 140/90 mm Hg, whether or not proteinuria is required, and the amount required for diagnosis remains controversial (Table 1.1) (3,5,6,7,8,9,10,11). This makes identifying PE cases and comparing research studies difficult. For the purpose of this study we have defined PE as per the Society of Obstetricians and Gynecologists of Canada (SOGC) guidelines (3) as: new or existing hypertension (BP 140/90 mm Hg) with one or more of the following i) new proteinuria (protein creatine ratio>30mg/mmol,>0.3g/day or >2+ dipstick) after 20 weeks gestation, or ii) one or more adverse maternal symptoms (headache, vision disruption, abdominal pain, severe nausea or vomiting, chest pain, dyspnea), maternal organ dysfunction (pulmonary edema, placental abruption), altered maternal lab results (elevated serum creatinine, AST, ALT, LDH, platelet count<100x109/L, or serum albumin <10g/L), or fetal morbidity (oligohydramnios, IUGR, and/or reversed end diastolic flow). We have opted not to include fetal mortalities in our cohort as stillbirth is a heterogeneous condition that can be caused by a number of etiologies (12).  PE is often subdivided into subtypes (13,14,15,16) in order to obtain groups that are more homogeneous. Early-onset PE (EOPE) and/or severe PE is more commonly associated with placental pathology and co-occurring IUGR (17,18,19). Molecular profiling of both the placenta and maternal serum reveal many different profiles of altered gene expression, DNA methylation (DNAm) or protein levels in PE pregnancies, suggesting that multiple subtypes of the disease may be present (20,21,22,23,24,25,26,27). LOPE, term PE, or mild PE, and EOPE, preterm PE, or severe PE are often used interchangeably, although these terms overlap, their definitions do differ. EOPE is defined as a diagnosis of PE prior to 34 weeks gestation, while LOPE is defined as a diagnosis after 34 weeks gestation (4,28). When defined by preterm PE, diagnosis of PE <37 3  weeks gestation, and term PE, diagnosis >37 weeks gestation, some LOPE cases become grouped with the EOPE cases. Mild and severe PE aim to define subtypes of PE based on severity, and as such are defined by maternal BP. Severe PE includes cases in which maternal BP> 160/110 mmHg, while mild PE is any case of PE where maternal BP sits between 140/90 mmHg and 160/110 mmHg, independent of gestational age(28). While both EOPE and LOPE are associated with maternal factors, there are some factors that are specifically associated with either EOPE or LOPE. Lisonkova and Joseph (2013) investigated 456,688 pregnancies and births, and found that nulliparity, young maternal age (<20 years), and diabetes mellitus was strongly associated with LOPE, while EOPE was more strongly associated with African-American race, chronic hypertension and congenital anomalies (29).  There are a number of factors that increase an individual’s risk of developing PE including: family history of or previous pregnancy with PE (4), advanced maternal age, ethnicity, and maternal obesity (Body Mass Index (BMI) >30) (30,31,32,33). Women with pre-existing chronic conditions, such as essential hypertension (34), diabetes, or vascular disease, develop PE more frequently than those that do not (35). Additionally, pregnancies that were conceived with donor sperm, donor oocytes or donor embryos have been associated with increased incidence of PE, likely due to immune intolerance (36). As one of the proposed pathogeneses of PE is inappropriate immune response, repeated or longer exposure to paternal antigens is thought to establish maternal immunity to paternal antigens, decreasing the risk of PE (37,38), although this has not been found conclusively (39). Fetal factors, such as multi-fetal pregnancies and fetal chromosomal abnormalities have also been known to co-occur with PE at a much higher rate (40,41,42). Risk factors and placental phenotypes associated with PE are outlined in Figure 1.1. 4  Currently treatment of PE is limited to antihypertensive medications to decrease maternal BP, and in some cases anticonvulsive medication (e.g. Magnesium sulphate) can be prescribed to treat eclamptic seizures (30), and if necessary, delivery of the baby. Whether or not these medications are good approaches to treating PE, as they could potentially harm the fetus long term, has been controversial. However, a recent study by Magee et al. (2015) found no difference in infant health a year after delivery in PE pregnancies treated with tighter regulation (target diastolic 85 mm Hg) compared to PE pregnancies under less tight regulation (target diastolic 100 mm Hg) (43). Managing both PE and IUGR requires balanced consideration of both maternal and fetal health by prolonging gestation to minimize poor neonatal outcome, but delivering early enough to avoid fetal demise and optimize maternal health outcomes. Daily aspirin has been suggested and is commonly used as a treatment to decrease the risk of developing PE in high-risk populations (44,45,46). However, decreased risk with aspirin prescription was observed when prescribing low dose aspirin to women at high risk of developing PE or IUGR; when expanding this finding to populations representing all risk levels, a large sample size was needed to observe a slight decrease in risk, for this reason low dose aspirin would only be beneficial in the high risk population (46,47,48).  While low dose aspirin has been found to reduce PE risk in high risk populations, how different health systems identify high-risk pregnancies differs. 5  Table 1.1. Clinical diagnosis criteria of preeclampsia worldwide.   **!May not be the most accurate diagnosis criteria for Africa as black women have generally high BP, more likely to conceive with chronic hypertension and more likely to develop hypertensive diseases during pregnancy. Fokom-Domgue et al (2014) suggests re-evaluating how to define PE in an African population.! 6  1.1.2! Intrauterine growth restriction Intrauterine growth restriction (IUGR), is defined as poor fetal growth due to a pathological mechanism. This is distinct from low birth weight, which is defined as birth weight <2500g (not corrected for gestational age) by the World Health Organization (WHO) (49), and small for gestational age (SGA) infants, which is defined as birth weight <10th percentile for the gestational age and fetal sex (50,51,52). IUGR is also variously defined as birth weight <10th percentile (50,53,54), birth weight of <3rd percentile(3,15,55), <5th percentile(56,57), or <10th percentile with other indications of poor growth (abnormal doppler findings)(15,55). Fetal growth restriction is a term now being used instead of IUGR in attempt to clarify and distinguish between SGA infants that are constitutionally small, but otherwise healthy, and IUGR infants that are small due to a pathological mechanism. For the fetal growth restriction definition, Mamelle et al. (2001) proposed an individualized model of birth weight taking into account gestational age, fetal sex, birth rank, maternal age, maternal height and maternal weight; this method tries to estimate an individual’s genetic potential of birth weight rather than setting a percentile threshold (58). Gardosi et al. (2000) also proposed the idea of individualized growth trajectories for fetuses using a model that incorporates fetal weight, gestational age, sex, maternal weight and height, parity and ethnicity(59). Distinguishing between a baby that is small but healthy, from one that has had growth limited due to extrinsic factors is challenging, but necessary in order for clinicians to appropriately manage pregnancy.   Measurements in utero that may indicate a risk of IUGR may include: ultrasound findings of fetal abdominal circumference (AC) <10th percentile for gestational age and sex or estimated fetal weight <10th percentile for gestational age (60), symphysis-fundal height (SPH) measuring 7  the size of the uterus (61), placenta ultrasound measures (length, thickness, fetal:placental weight ratio), and uterine and umbilical artery measures (notching, blood flow, pulsatility) (62). While these measures are useful and help clinicians best assess which fetuses are growth restricted vs constitutionally small, using thresholds defined by a proportion of the population will unavoidably result in a heterogeneous population of both pathologic and normal fetuses and will not include growth restricted fetuses that are above those thresholds. While in utero measures are needed to measure fetal growth trajectories over time and to prospectively identify pregnancies at risk of IUGR, definitions based on birth weight may be more helpful in research circumstances to compare results across studies. Unifying and stricter definitions may aid in identifying a more homogenous group of IUGR and associated causal mechanisms.  Risk factors of IUGR are similar to PE, and include i) pre-existing maternal conditions such as: chronic hypertension, diabetes and vascular disease (63); ii) previous pregnancy history such as a pregnancy diagnosed with IUGR or PE, iii) multi-fetal pregnancy,  iv) some fetal and/or placental  aneuploidies, and v) co-occurring PE itself (63). Figure 1.1 outlines both PE and IUGR risk factors. There are two different presentations of IUGR, i) asymmetrical IUGR, where the head is of normal measure (referred to as brain sparing), but the rest of the body is small and ii) symmetrical IUGR, where the fetus is overall small, but proportional. Pregnancies with asymmetric IUGR fetuses are at significantly higher risk of PE, and adverse neonatal and long term health outcomes (64,65). It has been hypothesized that asymmetric IUGR is the result of placental insufficiency, where reduced blood flow to the placenta and fetus result in favour of blood flow to the fetal brain. Whereas, symmetric IUGR is thought to be more commonly due to 8  teratogen or viral exposures during development, chromosome abnormalities, genetic syndromes or inborn errors in metabolism resulting in a proportional reduction in fetal size (66).   Both PE and IUGR are extremely heterogeneous conditions where many maternal, fetal and placental factors can contribute and influence the pathogenesis of disease (67,68,69,70). Asymmetric IUGR and EOPE are associated with abnormal placentation (66). In an attempt to adapt to the abnormal placentation, molecular characteristics in the placenta change and in many cases increased apoptosis and necrosis of placental cells which are then shed into maternal circulation, may result in an activated maternal immune response (71). It is unclear whether different subtypes of PE and IUGR are varying severities of the same condition (overall placental insufficiency), or whether they are distinct entities from one another. For example, it is unknown whether LOPE is a less severe phenotype of abnormal placentation and subsequent maternal immune response, or whether it is driven solely by maternal immune intolerance. Resolving this issue is essential to identifying sensitive and specific biomarkers to predict these conditions. Grouping together placental insufficiency conditions that are actually distinct entities will limit a study’s ability to identify true differences between the cases and controls, inhibiting our ability to identify the best candidate biomarkers. Subtypes with distinct entities will likely have different biomarkers that will predict each subtype, whereas subtypes that are different severities of the same condition could utilize the same panel of biomarkers. 9    Figure 1.1 Risk factors for preeclampsia (PE) and intrauterine growth restriction (IUGR). Risk factors lead to abnormal placentation and/or maternal immune intolerance, resulting in altered placental pathology and molecular profiles, giving rise to maternal and fetal phenotypes.   10  1.2! The placenta and its development 1.2.1! Placental structure The placenta is the organ involved in nutrient and gas exchange, blood flow to the fetus, hormone regulation and secretion, protection of the fetus from harmful xenobiotics, and blood cell production, once fully developed (72). The human placenta is made up of many tree-like structures, called chorionic villi. These chorionic villi are the sites at which nutrient and gas exchange take place. Distinct from some other placental mammals such as horses, cats and sheep, the human placenta is discoid in shape and is classified as hemochorial; meaning that the placental trophoblast cells come into direct contact with maternal blood, but the maternal and fetal vessels do not come into direct contact with each other (72,73). It is thought that the hemochorial placenta evolved as it possesses an advantage for more efficient nutrient transport resulting in a faster growth rate (74).   1.2.2! Placental development 1.2.2.1! Implantation Placental development occurs rapidly after fertilization. Cell division occurs, taking the single-cell zygote to a multi-cell blastocyst. The blastocyst consists of an inner cell mass (ICM) and a single-cellular trophectoderm layer that surrounds the ICM, and forms the blastocyst cavity. The trophectoderm layer will differentiate into the trophoblast cell lineages of the placenta and the ICM will contribute both to the fetal and extraembryonic tissues (75) (Figure 1.2). Between 6-12 days post-fertilization, the blastocyst implants into the uterine wall (76). The trophoblast cells, surrounding the ICM secrete metalloproteinases, which degrade the uterine lining (77). Trophoblast cells differentiate into mono-nuclear cytotrophoblast cells (73) which form column 11  structures that will become the chorionic villi. Cytotrophoblasts differentiate into either terminally differentiated multi-nucleated syncytiotrophoblast cells (by fusion of cytotrophoblast cells to each other) or extravillous trophoblast (EVT) cells. The multi-nucleated syncytiotrophoblast cells make up the outer layer of the chorionic villi and are involved in the initiation of placental development, invading the uterine wall (endometrium) and eroding the maternal tissues. Once the syncytiotrophoblast layer has initially invaded the endometrium, a subset of the cytotrophoblast cells will differentiate into EVTs, which migrate beyond the chorionic villi and syncytiotrophoblast cells to invade and remodel the uterine spiral arteries for increased blood flow to the placenta. Cytotrophoblast columns will then advance past the syncytiotrophoblast and connect to one another, forming the cytotrophoblast shell. Empty spaces between the cytotrophoblast columns are termed lacunar spaces, lined by the syncytiotrophoblast (78). The outer syncytiotrophoblast layer is in direct contact with the maternal blood, involved in the transfer of oxygen and nutrients to the fetus (73,79). The cytotrophoblast shell regresses and eventually forms the placenta basal plate where the chorionic villi are anchored. At 10-12 weeks gestation blood flow between the mother and the placenta is established. At this time, the fetus will fully utilize the placenta for nutrients and blood flow and subsequently increase its growth rate (73). 12   Figure 1.2. Cell differentiation of cell types that comprise fetal tissues and the placenta.13  1.2.2.2! Angiogenesis Placental angiogenesis, the establishment of placental and fetal circulations, is essential to fetal health and development. While differentiation of cytotrophoblast cells into either syncytiotrophoblast or EVTs is regulated by changes in both DNA methylation (DNAm, mitotically stable modifications made to DNA that can influence gene expression) and gene expression, oxygen tension plays a critical role as well (80,81). During early embryonic development, spiral arteries are plugged by cytotrophoblast cells to create the hypoxic conditions needed to promote cytotrophoblast proliferation (82). Hence, there is no direct connection between the mother and embryo, therefore there is little oxygen transport to the fetus. These hypoxic conditions promote cytotrophoblast proliferation and are essential for placental angiogenesis (83). Key angiogenic factors involved in placental vascularization are regulated by hypoxia induced gene regulation (84). At approximately 10-12 weeks gestation, the uterine arteries become unplugged, allowing for oxygen perfusion of the placenta through the differentiation of the cytotrophoblast cells into EVTs which remodel the uterine spiral artery (82). Once blood flow is established, maternal blood can deliver oxygen and nutrients to the developing fetus.   1.2.2.3! Nutrient transport During the first trimester, transport of nutrients from mother to fetus is histiotrophic through trophoblast phagocytosis (vesicle mediated) and/or endometrial glandular secretions. After 10-12 weeks, and the establishment of blood flow to the intervillous space, nutrient exchange is performed through the placenta (72). The primary mode of transport across the placenta is passive transport in which molecules move through the placental membrane based on the 14  concentration gradient (i.e. molecules move to the area with least concentration). This mode of transportation does not require any energy input however, migration across the placenta is dependent on the chemical properties of the molecule including polarity, hydrophilicity, size, and protein concentration in both the maternal and fetal blood streams. This mode of transportation is often used by hormones and oxygen exchange across the placenta (85). Active transport requires energy input in order for protein transporters to assist the molecules across the barrier. Essential nutrients are often transported through active transport as maternal concentrations of amino acids are often lower, therefore an active transport mechanism is required to move these nutrients across the placental barrier (86). Facilitated diffusion requires additional proteins to bind to the molecule and aid in its transport across the placenta; this mechanism is utilized by glucose due to its large size. Similar to passive transport, facilitated diffusion abides by a concentration gradient and does not require energy input. Vesicular transport is used to transport large molecules, unable to pass through the placenta by any other means. The molecules cross by being engulfed in microvilli that aid in their transportation (87). Modes of transport across the placenta and molecules that utilize them are listed in Table 1.2. Table 1.2. Methods of transport utilized by the placenta. Type of Transport Molecules Simple Diffusion/Passive transport Fatty acids, lipids, water: cross readily across the placenta Active Transport Most amino acids,K,Mg,Ca,PO4; requires energy to transport across Facilitated Diffusion Glucose and essential nutrients; protein mediated Rapid Diffusion Oxygen, C02, gases; rate limiting step is blood flow   15  1.2.2.4! Paracrine, endocrine, autocrine and immune functions in the placenta During pregnancy, the embryo and later fetus, with half its DNA foreign to the mother, must evade the maternal immune system to ensure its survival. There are several processes that are thought to be involved in the maternal-fetal interface to aid in immunotolerance. Paracrine autocrine, and endocrine signaling between the mother and the fetus is essential for immunosuppression (88). A variety of proteins are secreted from the uterine lining and uterine wall to aid in the implantation process and also inhibit trophoblast cells from invading too far into the maternal myometrium (89). Trophoblast cells invading the maternal tissues will secrete metalloproteinases and urokinase plasminogen activator to erode the uterine wall during the implantation and trophoblast invasion processes (89). Additionally, trophoblast cells will also secrete an array of immunosuppressive factors, and also lack major histocompatibility complexes and some human leukocyte antigen (HLA) to prevent an attack by the maternal immune system (89). Shedding of placental trophoblast cells into the maternal circulation may be another process by which the maternal immune tolerance is achieved (89).  1.2.2.5! Placental development in preeclampsia and intrauterine growth restriction The pathogenesis of PE and IUGR remains relatively unclear, however current hypotheses revolve around poor placental invasion and remodeling of the uterine spiral artery (4,30,67). This can result in decreased blood flow leading to hypoxia in the placenta and fetus. Hypoxia is a normal phase of development (67,90), however, prolonged hypoxic state results in rapid reoxygenation which inhibits trophoblast differentiation leading to a reduction of syncytiotrophoblast cells responsible for hormone production and nutrient regulation required for pregnancy maintenance as well as fetal and placental growth (91,92,93). Exposure to a prolonged 16  hypoxic state, and subsequent reoxygenation, can cause irreparable damage and result in increased shedding of microparticles into the maternal blood stream; this may produce an inflammatory response leading to the clinical phenotype of PE (94,95). Abnormal placentation may also be due to a negative response of the maternal immune system to the paternal antigens. An inflammatory response occurs when the female is exposed to paternal antigens. This response may be crucial to allow the maternal immune system to process paternal antigens and inhibit inflammatory responses against any future conceptus (30,33,96).  1.3! Current screening for placental insufficiency 1.3.1! Imaging tools 1.3.1.1! Ultrasound Currently, ultrasound remains the standard approach to placental and fetal imaging during gestation. This is due to its minimal cost, safety (non-invasive and no radiation), and that information on the health of the placenta can be obtained as early as eight weeks of pregnancy (97). The Doppler ultrasound was developed to assess appearance and location of the placenta, but also measure blood velocity within the placenta. Therefore technicians can gauge fetal: placental perfusion as well as vascular impedance (97). However, ultrasound is limited by the angles in which the ultrasound wand can visualize the placenta. Placental magnetic resonance imaging (MRI) and positron emission topograph (PET) are able to overcome this particular obstacle.   17  1.3.1.2! Magnetic resonance imaging and positron emission topograph Although not routinely used to assess the placenta, MRI and PET are exceptional tools to visualize both placental appearance and function. Overcoming the limitations of ultrasound, MRI and PET can produce images from all axes of the placenta (97). MRI has such fine resolution, that it can estimate blood flow within localized regions of the placenta (97). Commonly, MRI requires a contrast agent in order to visualize the placenta, and few studies have looked at the safety of using these agents in pregnant humans (98,99). For this reason, placental MRI is used mostly as a confirmatory tool in cases where the benefits of doing the imaging outweigh the risks; these are usually cases where it is hard to visualize the site of implantation, or hard to measure low velocity blood flow with great concern for the outcome and wellbeing of the fetus (100,101). However, recent studies have used a phase-contrast MRI, not requiring a contrast agent to measure blood velocity in the field (102). This methodology can potentially be used to measure blood velocity in the placenta. Similarly, PET imaging requires radioactive dye which can cross the placental barrier, thus it has not been recommended for use in humans (100). More recent studies have suggested that the amount of radioactivity would not be harmful (103), while others push for more animal trials to better understand radiation dose effects on the fetus (104).   Imaging the placenta during gestation can give us information as to the placental function, and a woman’s risk of developing complications due to placental insufficiency. Currently, a large meta-analysis using uterine artery Doppler ultrasound data on 55, 974 first trimester pregnancies produced a sensitivity and specificity of detecting EOPE of 47.8% and 92.1%, respectively. Similarly, sensitivity and specificity of detecting symptoms indicative of early onset fetal growth restriction was 39.2% and 93.1%, respectively; and the sensitivity and specificity of detecting 18  symptoms of any onset of PE (26.4% and 93.4%, respectively) or IUGR (15.4% and 93.3%, respectively) (105).While current imaging techniques can be fairly certain in the high risk pregnancies they identify, they are missing a large proportion of high risk pregnancies. Additionally, other caveats remain to the efficiency of detecting high risk women. Although poor implantation or infarctions visible in the placenta would indicate a high risk pregnancy, in some cases the placenta is able to overcome these obstacles (106). Additionally, placental lesions and infarctions are commonly seen in healthy pregnancies especially at later gestational ages (107).  Placental imaging as a whole is limited by cases in which abnormal blood flow to the placenta and/or placental appearance anomalies are detectable to call a high risk pregnancy. Many pregnancy complications will arise with no indication of placental abnormalities on ultrasound (106). To improve our ability to detect women at high risk of developing many types of pregnancy complications, due to many pathogeneses it has been suggested that combining ultrasound findings with maternal health history and maternal serum markers will increase sensitivity and specificity. Rolnik et al. (2017) (46) suggest the use of  maternal factors, mean arterial pressure, uterine-artery pulsivity index, and maternal serum pregnancy-associated plasma protein A and placental growth factor, as used in Akolekar et al. (2013) with a detection rate of 76.6% of preterm PE and false discovery rate of 10%(108), however this is not significantly better than the results from using just Doppler ultrasound (105). Improving the identification of high risk pregnancies will allow for preventative treatment to pregnancies in need and aid in optimizing our identification of women who will subsequently develop complications such as PE and IUGR (109).   19  1.3.2! Maternal serum markers Currently, women in Canada are offered maternal serum screening during their first and second trimesters of pregnancy to identify pregnancies at risk for fetal chromosome abnormalities (specifically trisomy 18 and 21), and neural tube defects. The maternal serum screening panel consists of five protein biomarkers: Pregnancy associated plasma protein A (PAPPA), human chorionic gonadotropin (hCG), alphafetoprotein (AFP), estriol, and inhibin A (INHA), all of which are measured by blood tests (110,111,112). Abnormal levels of these markers that are not explained by chromosomal abnormalities or fetal anomalies may be indicative of obstetrical complications such as: placenta accreta, increta, or percreta, spontaneous abortion, preterm birth, IUGR, PE, intrauterine fetal demise, or placental abruption. Studies developed in the United Kingdom have shown much promise in using maternal screening markers to predict adverse pregnancy outcomes, particularly PE and IUGR (109). However, there are concerns as to whether these results are generalizable to other populations and whether PAPPA and placental growth factor (PLGF) levels predict high risk women more than a basic personal and family history (113). New approaches are shifting the focus from ultrasound and protein biomarkers in maternal serum to the presence of cell free DNA, RNA and miRNA in maternal blood (114).  1.4! Molecular studies of the placenta in placental insufficiency 1.4.1! Chromosomal associations One of the few genetic associations known to increase incidence of PE and IUGR is the presence of trisomic cells in the placenta (42). These could be cases of constitutional trisomy (trisomic cells also present in the fetus), or confined placental mosaicism (CPM) (trisomic cells confined to placenta). CPM occurs in up to 2% of pregnancies ascertained through chorionic villi 20  sampling (115) and is associated with increased incidence of IUGR (116). CPM involving trisomy 16 (CPM16) has been associated with increased incidence of PE (117). DNAm profiling showed that CPM16 placentas had an overlapping profile with EOPE placentas (118). Few genes located on chromosome 16 have been associated with PE, however chromosome 16 harbors several genes involved in placental development (ex: CDH1 and ERK1, involved in trophoblast differentiation and invasion, respectively) (119,120). It is hypothesized that the extra chromosome 16 results in shallow trophoblast invasion into the uterine wall, resulting in PE. However, the mechanisms of how this occurs remains unknown (118).   1.4.2! Genetic variants Genetic studies investigating the heritability component of PE have estimated heritability to be between 0.22-0.54 (121,122,123). While there are gene mutations that appear to segregate within families along with PE development (124,125,126), there has only been one genome-wide study that identifies and validates a single gene variant associated with increased susceptibility to PE (127). PE and IUGR are multifactorial etiologies, where numerous low penetrant genetic variants contribute small risk of PE and/or IUGR development, combined with environmental and maternal health exposures, result in disease manifestation (128). Many of the studies take a candidate gene approach, investigating single nucleotide polymorphisms (SNP) in genes involved in pathways known to play a role in placental insufficiency. This approach reduces the number of multiple test corrections needed, and increases the power of the study. However, it makes an assumption of causative pathways based on the current understanding of the pathogenesis of PE and IUGR, which is still relatively unclear.  Allelic variation in genes acting within the angiogenic pathway (129), solute carrier pathway (130), endoglin pathway (131), and 21  inflammatory and immune response pathways (132,133,134), have been identified. However, results from these candidate genes studies have low reproducibility rates in other cohorts (134,135). Major contributors to the low reproducibility rate may be different subtypes of PE and IUGR and how they are defined, different variants being present in different populations, and that we expect low effect sizes, and the small sample size used in many studies is severely underpowered to detect such minute differences (136).   In recent years, several groups have performed genome-wide association studies (GWAS) in an attempt to uncover some of the missing heritability of the disorder (137-144). While this approach allows for investigating all variants without an a priori hypothesis, the number of variants being assessed results in many multiple test corrections. For this reason, massive sample sizes are needed to have enough power to detect variants with small effect sizes. Early studies from Australia, Finland, Holland, and Iceland all identified PE associated risk loci in their cohorts. However, these results have not yet been validated in independent cohorts. A large-scale effort to identify PE-associated variants in the Netherlands (InterPregGen) consists of a discovery cohort of 57,600 maternal serum samples, of which 11,600 samples had diagnosed PE (142). The consortium also had an additional 14,000 samples, of which 9,200 had diagnosed PE in which to validate any potential findings. Recently, InterPregGen reported a genetic risk to PE development when individuals possessed a variant within FLT1 (127), a gene and subsequent protein that has been consistently associated with PE (129,145,146,147). However, this particular variant is very common and does not fully explain the heritable component of PE.   22  1.4.3! Telomere length variation Telomeres are repetitive sequences of DNA, located at chromosome end to protect them from end-to-end fusion and DNA damage.  As DNA polymerase cannot synthesize DNA in the 3’-5’ direction, telomeres are reduced in length with every cell division in the absence of telomerase activity. Oxidative stress is a known mechanism that increases DNA damage and results in shortened telomere length and cell death (148) and is thought to be involved in PE and IUGR (4). For this reason, altered telomere length in the placenta has been proposed as a marker of PE and IUGR and has been investigated by several studies. Some studies reported shortened telomere length in PE and IUGR placentas, while other were unable to reproduce their results (149,150,151,152,153).  The lack of reproducibility may be explained by differing models; some studies include fetal sex and gestational age as confounding factors and others do not (149-153), while some studies combine all PE together and others subdivide cases into EOPE and LOPE (149-153).  1.4.4! Transcriptomic alterations 1.4.4.1! Altered gene expression in preeclampsia and intrauterine growth restriction There are numerous studies investigating altered gene expression in placentas from pregnancies complicated by PE and/or IUGR (20,21,147,154,155-161). Genes involved in cell proliferation, cell structure, lipid metabolism, transport, and angiogenic factors regulated by hypoxia have been reported overexpressed in PE (147,157,162), while genes involved in cell proliferation immunity, cell adhesion, and cell motility have been reported under expressed in PE (147,154,163). Interestingly, Sitras et al. (2009) compared gene expression between EOPE and LOPE and found that altered expression of genes involved in oxidative stress, inflammation and 23  endoglin signalling were more associated with EOPE than LOPE (147). Transcriptomic profiling in PE, IUGR, and control placentas has revealed distinct clusters of transcriptomic profiles, indicating multiple subtypes of PE (20,21). Similar to placental DNAm, gene expression is highly variable across the placenta (164); as such, many studies will opt to sample numerous sites and pool the DNA and/or RNA together to give a more accurate representation of the placenta as a whole. Although gene expression alterations are more intuitive to understand in the context of gene regulation, gene expression is dynamic over gestational age and the alterations one is observing at the time of delivery may not necessarily reflect the alterations present at an earlier time point, which is when these alterations would be used to predict a future adverse outcome. Additionally, gene expression can differ between the mode of delivery (i.e. vaginal vs caesarean) (165,166), and mRNA rapidly degrades after birth, making sample processing time a confounding variable (164).   1.4.4.2! Altered miRNA profiles in preeclampsia and intrauterine growth restriction As microRNA (miRNA) are important to gene regulation and are more stable than their mRNA counterparts, investigations into miRNA profiles in relation to PE and IUGR have been very popular. MiRNAs are involved in many processes during placental development, including: trophoblast invasion (miR-195, miR-21), migration (miR-376c, miR-378-5p), angiogenesis (miR-16, miR-29b), trophoblast cell differentiation (miR378a-5p,miR376c,miR-141),and apoptosis (miR-29b) (167). During pregnancy, miRNA, as well as other nucleic acids are released into maternal circulation, through cellular apoptosis and necrosis, or vesicle mediated transport (114). A number of circulating miRNA, both specific and non-specific to the placenta, have been proposed as potential biomarkers in PE prediction (167); specifically miR-210 has 24  been associated with PE and show promise in having clinical utility (168). No miRNAs have been proposed for potential prediction of IUGR yet (114). No miRNA or panel of miRNA have high enough sensitivity or specificity to be utilized in the clinic as of the publication of this dissertation.   1.5! DNA methylation DNA methylation (DNAm) is a type of epigenetic mark. Epigenetics is defined as the study of mitotically stable chemical modifications made to DNA or the proteins around which DNA binds that do not change DNA sequence but influence gene regulation (169). Epigenetic marks include histone modifications, chromatin organization, and DNAm. DNAm is the most widely studied epigenetic mark, as it is relatively easy to measure, and the relationship between DNAm and gene expression has been extensively studied (170). DNAm is the addition of a methyl group by DNA methyltransferases (DNMT), most commonly to the 5-prime position of a cytosine base in cytidine-phosphate-guanosine (CpG) dinucleotides (171). During DNA replication DNMT1 recognizes hemimethylated CG sites in newly replicated DNA in which the template contains a methyl group and the product does not. DNMT1 is responsible for adding a methyl group to the complimentary strand in order to maintain the DNA methylation profile through cell division (172). In addition to DNMT1, DNMT3a and DNMT3b also have the ability to add methyl groups to CG sites. However, DNMT3a and DNMT3b do not recognize hemimethylated states; rather they can de novo methylate previously unmethylated CG sites and play a crucial role in establishing the methylation profile during embryonic development (173).   25  In some regions, CpGs are clustered together in “CpG islands”, regions of high CpG density which are generally found at promoters (174). How DNAm influences gene expression is dependent on genomic location. Many studies show correlation between DNAm and gene or protein expression, with DNAm in a promoter region generally correlating with reduced or no gene expression or protein expression (175,176).  Methyl groups bound to DNA can repress gene expression by either actively preventing transcription factors from binding to the promoter or to enhancers, or recruiting methyl-binding proteins that compete with transcription factors in promoter binding. Although DNAm at promoter regions in the genome is most commonly studied, DNAm present outside of promoters can also influence or be associated with gene expression (177). For example, many DNAm changes occur in CpG island shores, the region directly adjacent to the CpG island. In these regions, decreased methylation is often associated with DNA polymerase activity and gene transcription (178).   1.5.1! Influencing variables of DNA methylation 1.5.1.1! Sex specific DNA methylation DNAm differs between males and females (179,180). DNAm plays a role in X-inactivation, the silencing of the second X chromosome in XX female individuals, while X-inactivation does not occur in XY males (181). This process accounts for only some of the differences in DNAm patterns between male and females. Sex differences in DNAm has also been observed on autosomes, suggesting biological and epigenetic differences between males and females extend beyond the X and Y chromosomes (180). The reason for these differences remains largely unknown, however it has been postulated that DNAm differences between the sexes may be influencing gene expression within hormone secreting pathways (182).  Additionally, studies 26  have shown altered DNAm patterns between males and females in response to environmental exposures and disease (182-185). For these reasons, when investigating differences in DNAm in association with a variable of interest sex should either be adjusted for within a statistical model or the study groups should be matched for sex to reduce the chance of identifying changes in DNAm that are actually associated with sex rather than the variable of interest.  1.5.1.2! Environmental exposures The term environmental exposures encompasses many factors, such as: smoking, air pollution, diet, teratogen, radiation, and general stress, to name a few. In evidence of environmental influences on DNAm, monozygotic twins, who share their DNA sequence, exhibit different DNAm patterns suggesting external influences on DNAm (186-188). This could just be CpG sites that are influenced by stochastic factors or not maintained; however, there is a wide array of literature on alterations in DNAm in association with many environmental exposures, such as maternal smoking. An area that has gained much interest in research of DNAm and the placenta, and factors that influence DNAm in the placenta is the developmental origins of health and disease (DoHAD) hypothesis, whereby in utero exposures influence long term health outcomes in adult life (e.g. obesity, diabetes and cancer)(189-191).   1.5.1.3! Disease state Alterations in DNAm have been widely reported in many diseases using both specific candidate gene driven approaches and genomic-wide approaches (22,192,193,194,195,196). While disease outcomes have often been studied in the context of gene mutations, in most cases, those mutations do not account for all subtypes or incidences of the disease. Alterations in DNAm (and 27  other epigenetic marks) may explain the missing mechanisms of disease not associated with genetic sequence (197,198).  As such, DNAm profiling has been used to obtain insight into disease pathogenesis and to identify candidate biomarkers (199,200,201). DNAm biomarkers are already being utilized in the clinic, with the use of Septin 9 (SEPT9) promoter DNAm being used to diagnose colorectal cancer (202). Further studies are underway to characterize the DNAm profiles of many different cancers, and to utilize the circulating cell-free DNA in blood to non-invasively detect cancer, its location, and prognosis of the patient (203,204,205,206). These methods are also being used in complex diseases where genetic profiling has yielded few associations (e.g. Alzheimers, Parkinson’s and Heart disease) (207,208,209).   1.5.1.4! Tissue-specific DNA methylation DNAm varies widely across cell types (210,211). This is thought to be related to the epigenetic commitment of cell fate, where epigenetic marks, such as DNAm, fine-tune gene expression, specifically at transcription factors, to fixate cell lineage and fate (212). As such, different cell types or tissues retain unique epigenetic signatures (210). In samples where there are mixed cell type populations (e.g. blood), a change in DNAm could either mean an average change in DNAm across all cell types in the sample or an alteration in the ratio of different cell types. While making this distinction may not be important for the aim of developing biomarkers of disease, it could be relevant to distinguishing an epigenetic change from a cell type-ratio change. Algorithms have been developed to correct for cell type composition in studies where DNAm changes may be influenced by altered cell populations between the cases and controls. In well-studied samples, where cell sorting is relatively easy (e.g. blood and brain), reference based methods for cell-type correction have been developed (213,214). For tissues, where reference 28  based methods have not been developed, a reference-free method has been developed to try to mediate the effects of cell composition in complex tissues (215). However, as changes in DNAm differ widely across tissues, and constitute the most variation in data, the reference-free method to correcting for cell composition assumes that the first few principal components in the data are due to cell composition, and removes them. This may remove some variance associated with the variable of interest, and does not take into account that some cell types will be more related to each other than others, meaning there still could be variation in the data due to cell composition. Additionally, in some situations, alteration in cell composition may be associated with the disease pathogenesis itself. If one were to correct for cell composition, then the variance due to pathology may be significantly reduced or eliminated.  In the placenta, no reference method has been developed to assess alterations in cell composition. While knowing whether DNAm alterations are an average change across all cells sampled, or whether DNAm alterations represents a change in cell composition does need to be addressed and will give insight to the pathogenesis of PE and IUGR, it will not affect the utility of DNAm as a biomarkers for pregnancies at risk of developing PE and/or IUGR. DNAm marks and biomarkers in general do not need to be directly involved in the pathogenesis to be a good indicator of placental insufficiency or the variable of interest.  1.5.1.5! Age associated DNA methylation Individual/patient age is a factor widely known to influence DNAm (216,217). Numerous pan-tissue or tissue specific epigenetic clocks have been developed (218-221), try to predict chronological age. The biological meaning of CpG sites that correlate with chronological age 29  remains unclear. However the ‘epigenetic age’ of an individual has been associated with time to death (222,223), cancer (223-225), chronic stress (226), degenerative disease (227), and obesity (228). Similar to changes in DNAm observed in association with chronological age, DNAm in the placenta changes over gestational age (229,230). For this reason, gestational age should be accounted for in statistical models or have the study groups matched for gestational age.   1.5.2! The placental methylome Human development starts with a single cell zygote. This cell begins in a totipotent state dividing and differentiating into many different cell types, contributing to the fetal tissues, placenta and extraembryonic membranes (231). As cells each contain the same DNA sequence, a mechanism must exist in order for cells to differentiate into different cell types. Alterations in epigenetic marks, specifically DNAm, distinguish different cell or tissue types. Very early in development cells undergo ‘epigenetic reprogramming’, a process by which the epigenetic marks are erased from the genome and re-established as specific cell or tissue profiles (232).   The DNAm profile of the placenta is quite different to other somatic tissues. There are many partially methylated domains (PMDs)(233) and imprinted genes within the placenta(234). This results in a trimodal distribution of DNAm measures, where many CpG sites have an average 50% methylation resulting in a third peak, as opposed to the bimodal distribution of somatic tissues, where most sites are either on average 0% methylated or 100% methylated(235). The placenta is globally hypomethylated, with DNAm in the chorionic villi showing up 25% less DNAm than fetal tissues(236). Additionally, the placenta also contains partially methylated domains (PMD); these are large genomic regions (>100kb) with reduced DNAm interrupted with 30  regions of higher DNAm(233). Retrotransposable elements have been reported to be hypomethylated in the placenta(237). The placenta is highly heterogeneous and DNAm is incredibly variable across a single placenta and between individuals (238). This variability can be attributed to a number of factors known to influence DNAm.  1.5.3! DNA methylation in preeclampsia and intrauterine growth restriction Alterations in DNAm in both PE and IUGR have been reported with both candidate gene targeted approaches and genome-wide approaches (15,22,23,50,51,52,53,54,56,118,196,221,239-270) (Supplementary Table A.1). Many genome-wide DNAm studies show alterations in genes associated with oxidative stress, cell proliferation, inflammation and immune response (53,241,243,249). However, not all potential confounding factors, such as fetal sex and gestational age, are assessed in many of these studies. Therefore many of the PE/IUGR ‘hits’ may actually be artifacts of gestational age or fetal sex.   Within the literature, there is an inconsistent definition of what is useful to use as a control, when comparing to PE and IUGR. Some studies will compare preterm PE and IUGR to preterm control placentas (22) ; however, preterm birth are still abnormal, meaning that changes in DNAm may actually be changes related to the cause of the preterm birth group rather than the PE or IUGR group. To minimize this effect, using a variety of preterm birth aetiologies (ex: spontaneous premature rupture of the membranes, incompetent cervix or chorioamnionitis) will mitigate the likelihood that a change in DNAm is due to a singular cause of preterm birth. Other studies opt to compare PE and/or IUGR to healthy term placentas (243,244,254), however, this often results in the pathology and control groups being confounded by gestational age. Although 31  DNAm alterations are reported in both PE and IUGR, few studies validate their findings in an independent cohort, and reproducibility rate and prediction rate remains relatively low (240). Validation and reproducibility will be discussed in detail in the discussion.  1.5.4! Measuring DNA methylation There are numerous methods that have been developed to measure DNAm both across the genome and in smaller more targeted approaches. Which method to use, will depend on the experimental question and the genomic region of interest. Some considerations to take into account when choosing a platform are: cost, amount of input DNA required, if any pre-treatment is necessary and if that will result in any DNA loss, technical variability, how many CpGs are covered by the platform, whether there is any bias in the genomic regions covered and the experimental question.   1.5.4.1! Genome-wide DNA methylation measures DNAm microarrays are a commonly used method to measure DNAm across the genome. They cover a small percentage of total CpG sites within the genome, and are enriched at high CpG density regions of the genome (271).The Illumina microarrays will be discussed in more detail in Chapter 2. Reduced representation bisulfite sequencing (RRBS) involves methylation sensitive enzymatic digestion and size selecting for small fragments (CpG dense regions). This method covers a higher proportion of CpG sites compared to microarray (272). Methylation DNA immune precipitation (MeDIP) sequencing interrogates DNAm at methylated regions, as antibodies specific to 5-methylCytosine are used to isolate methylated fragments of DNA (272). While this reduces the number of sites being assessed, it is difficult to determine if no reads are 32  reported, if the result is truly no methylation or whether it is a technical artifact (no reads) (273). Whole genome bisulfite sequencing covers all sites in the genome, but this generates a lot of data where identifying significant differences is hindered by multiple test correction and the relationship between DNAm and gene expression can be less intuitive in some genomic regions (274).  Measuring DNAm at repetitive regions (e.g. LINE1 and ALU) have been used as surrogates for global DNAm levels. However, assays measuring global DNAm appear to be assessing different combinations of these repetitive elements and global DNAm levels do not correlate between the different assays or elements (237). The coverage of DNAm measures and the affordability of the techniques vary widely. Which technique to utilize for a given study will depend on the research question, and goals of interest.  1.5.4.2! Candidate gene DNA methylation measures Experimental questions investigating targeted sites or a small subset of sites will often use bisulfite pyrosequencing which can cover up to 350bp with one sequencing primer, and DNAm at numerous CpGs within those base pairs(275,276). The accuracy, efficiency, high correlation to other techniques, small amount of input DNA required and low cost per sample make this an ideal approach for those interested in studying candidate regions within the genome(275,276).  Other methods utilizing methylation specific enzymatic digestion and amplification (PCR), custom microarray chips, targeted bisulfite sequencing are also used.   1.6! Research hypothesis and objectives Previous students in the Robinson lab have investigated placental DNAm for both its underlying biological properties and its association with disease. Novakovic et al. (2011) utilized the 33  Illumina 27K array platform to profile DNAm patterns in placental and fetal tissues over gestational age (277). Yuen et al. (2010) also identified promoter hypomethylation in EOPE placentas (23). Following up this work and expanding the number of CpG sites assessed to >480,000 sites across the genome, using the 450K array, Blair et al. (2013), identified enhancer hypomethylation in EOPE placentas (22). Blair also identified overlapping DNAm alteration in trisomy 16 placentas, which confer a higher risk of developing PE (118). In this dissertation genetic and epigenetic profiles in EOPE, LOPE, and IUGR were assessed. I hypothesized that EOPE, LOPE and IUGR were distinct entities from one another and would therefore have unique genetic and epigenetic profiles that could potentially be utilized for biomarker development. The research objectives of this dissertation were to i) assess the utility of DNAm use in biomarker development, ii) reassess and characterize the DNAm profiles of EOPE, adding a few new samples and new bioinformatic analysis and expand the analysis to encompass LOPE and IUGR, and iii) to assess the utility of placental telomere length as an indicator of placental stress during gestation. To complete these objectives, I undertook the following studies: (i)! A comparison and correlation between DNAm in the term placenta to protein levels in maternal serum during the first, second and third trimesters of pregnancies.  (ii)! A characterization of the DNAm profiles associated with EOPE, LOPE, and IUGR compared to control placentas. DNAm profiles were further refined with the use of an independent validation cohort obtained from Dr. Brian Cox’s lab at the University of Toronto.  (iii)! An assessment of telomere length dynamics in the placenta over gestational age and whether telomere length is altered in EOPE, LOPE, and IUGR compared to control placentas, taking into account gestational age and fetal sex.  34  These studies resulted in further insight into the relationship between EOPE and LOPE and ideas as to how DNAm can be utilized to improve PE prediction.  35  Chapter 2:!Methods In this dissertation, placental sampling and genome-wide DNAm measurement using the Illumina Infinium HumanMethylation450 array (450K) platform were performed the same way through multiple data chapters. The methods of these protocols are outlined in this chapter.   2.1 Case ascertainment  Ethics approval was granted by the University of British Columbia and BC Women’s and Children’s Hospital Research Ethics Board (H04-70488) for our EPIC study (Epigenetics of the Placenta in Complications of Pregnancy). This study includes two groups: PM (database name) cases with informed consent and full clinical details and PL (database name) cases obtained predominantly from the embryo-pathology lab and have been de-identified with minimal clinical details. Maternal clinical details include: self-reported ethnicity, pre-pregnancy body mass index (BMI), obstetric history, substance use in pregnancy, maternal health status (diabetes, chronic hypertension, other complications), medications and other supplements. Fetal and pregnancy details include: mode of delivery, infant sex, birth weight, and gestational age at delivery. We calculated birth weight defined by its’ standard deviation (SD) relative to normative values by gestational age and fetal sex (278).   In cases of preterm birth, known etiologies such as: spontaneous premature rupture of the membranes, spontaneous preterm birth, incompetent cervix, preeclampsia, and IUGR are recorded. All data is in a REDCap database hosted at the BC Children’s Hospital Research Institute affiliated with the University of British Columbia.  36  2.1! Placental sampling Chorionic villi samples, approximately 2 cm3, were taken from the fetal side of the placenta from four distinct placental cotelydons (164). One site was located near the umbilical cord and one near the periphery of the placental disk. Other sites sampled were located approximately halfway between the cord and placental edge and well-spaced across the placenta. The amnion and chorion (fetal membranes) were removed from the placental surface of each sample site and whole chorionic villi were dissected with tweezers, followed by thorough sample washing to remove potential contamination of maternal blood. DNA was extracted from each sample by a standard salting out method, and then pooled together in equal amounts. The salting out method involves the use of high concentration of salts to precipitate proteins to the bottom of the solution. The DNA, within the supernatant is then precipitated and extracted using isopropanol. DNA integrity was assessed in all samples upon extraction. DNA purity and concentration were assessed using the Nanodrop 1000 spectrophotometer (ThermoScientific, Wilmington, DE, USA). Multiplex ligation-dependent probe amplification (MLPA) was used to screen all placentas for chromosome abnormalities.  2.2! Clinical definitions of preeclampsia and IUGR Preeclampsia (PE) was defined according to the Society of Obstetricians and Gynecologists of Canada (SOGC) criteria as gestational hypertension (existing or new) with one or more of the following i) new proteinuria (protein creatine ratio>30mg/mmol,>0.3g/day or >2+ dipstick) after 20 weeks gestation, or ii) one or more adverse maternal symptoms (headache, vision disruption, abdominal pain, severe nausea or vomiting, chest pain, dyspnea), maternal organ dysfunction (pulmonary edema, placental abruption), altered maternal lab results (elevated serum creatinine, 37  AST, ALT, LDH, platelet count<100x109/L, or serum albumin <10g/L), or fetal morbidity (oligohydramnios, IUGR, and/or reversed end diastolic flow), or HELLP syndrome with or without hypertension(279). However, we did not consider hypertension with stillbirth a PE diagnosis, as we felt stillbirth could be due to a number of etiologies (12). PE was subdivided into early-onset PE (EOPE), defined as a diagnosis of PE before 34 weeks gestation, and late-onset PE (LOPE), defined as a diagnosis of PE at or later than 34 weeks gestation (280).   IUGR was also defined following SOGC criteria (278) as birth-weight <3rd percentile accounting for both fetal sex and gestational age, or birth-weight <10th percentile with additional clinical findings indicative of poor placental function such as: absent or reversed end diastolic velocity on Doppler ultrasound, or oligohydramnios. Criteria for exclusion were multi-fetal pregnancies, and fetal chromosomal abnormalities. Controls were selected based on absence of any criteria listed above and a placenta with no observable pathology. In some cases we define a normotensive IUGR (nIUGR) group, which are cases diagnosed as IUGR but after excluding maternal hypertension and/or preeclampsia.  2.3! DNA methylation microarray- Illumina Infinium HumanMethylation450 array Microarrays are widely used to evaluate DNA methylation (DNAm) at CpG sites across the genome, typically enriching for CpG dense regions, promoters and enhancers (271,281). For this dissertation, genome-wide DNAm was measured using the 450K platform, measuring 485,512 CpG sites across the genome. Relative to whole genome analysis such as whole genome bisulfite sequencing (WGBS), microarrays have a lower cost per sample and can limit the number of multiple test corrections needed, and give information on a site by site basis. Limitations of the 38  microarray approach are i) there is less complete coverage of the genome and ii) a higher quantity of input DNA is required compared to bisulfite sequencing techniques, which measure DNAm across the entire genome, not just at a subset of sites (272).   750ng of DNA purified using Qiagen blood and tissue kit was bisulfite converted using the EZ DNA Methylation Kit (Zymo Research, Irvine, USA). Samples were PCR amplified before being hybridized to the microarray chip as per the manufacturer’s protocol, and microarray chips were scanned by a HiScan 2000 (Illumina).  On the Illumina Infinium HumanMethylation450 (450K) array, each microarray chip can analyze 12 samples concurrently. The surface of the microarray contains thousands of beads, with many probes (single stranded DNA sequence, complementary to a genomic region) attached. These probes will complementarily bind to the target genomic location of interest and measure DNAm at its target site. DNAm is measured based on genotyping of C/T SNPs following bisulfite conversion; whereby unmethylated cytosines are converted to uracil, and ultimately replaced with thymine upon PCR amplification (271). Each fragment of DNA bound to a bead will be either methylated or unmethylated, a C or a T genotype. The DNAm measure reported is an average of a particular CpG within an entire sample. On the 450K array there are two types of probes, differing by the chemistry of their design and the genomic regions which they target. Type I probes utilize a 2-bead system, a methylated and unmethylated bead, where the probe sequence differs by the C/T genotype of the target CpG. The type II probes utilize different colour fluorescence (red and green) to label the methylated and unmethylated fragments. Fluorescent intensity is translated into DNAm values for each CpG site. Type I probes are utilized to cover high CpG density regions, while type II probes are used to cover lower CpG density regions (271). While this is needed in order to obtain 39  DNAm measures for different regions of the genome, it results in DNAm values being confounded by the probe type used to assess specific CpGs. Normalization is used to correct for probe type and other technical artifacts in the data. While there are a variety of normalization methods developed for the 450K array (e.g. Functional normalization (282),subset-within array normalization (SWAN)(283), Peak based correction (284), β-mixture quantile normalization(285)), some of these methods do perform better than others depending on the sample tissue type and the magnitude of change that is expected between the variable of interest and control (286,287,288).   In chapter 3, DNAm microarray data from Blair et al. (2013) was used to identify the candidate CpG sites. This data differs from the DNAm analyses in chapters 4 and 5, as it uses subset within-array normalization (SWAN) (22). In chapter 4 and 5, raw data (IDAT Files) were read into R statistical software, version 3.2.4, where functional normalization (282), background subtraction and colour correction were performed. Functional normalization utilizes the 848 control probes on the array to estimate changes in DNAm that are due to technical effects (e.g. sample placement on microarray, whether between microarrays in study or on the array itself) (282).  In chapter 5, batch effects were corrected using ComBat (289). However, after publication of chapter 5, I realized that because our samples’ gestational ages were not evenly distributed across batches, I increased the risk of introducing spurious finding in the data (290). For this reason, chapter 4 does not correct for batch effects and instead utilizes an independent validation cohort to identify true differentially methylated sites, as batch effects are unlikely to affect two cohorts the same way.  40  The discovery cohort is used in both chapters 4 and 5, while the validation cohort is used only in chapter 4. In both chapters, bad quality probes, those that had a missing beta value (value on <3 beads on the array) in > 5% of samples or a detection p-value (background intensity; used to determine if intensity values are likely true) <0.01 were removed from the analysis (Discovery N=1,402, Validation N=1,115). To minimize fetal sex effects, probes on the X and Y chromosomes (Discovery N= 11,648, Validation N=11,302), probes that cross hybridize to the X and Y chromosomes (Discovery N= 11,412, Validation N=10,734), and probes containing a SNP at the CpG of interest were also removed (Discovery N= 19,957, Validation N= 20,398) (291).  This left 440,093 CpG sites for the analysis in the discovery cohort and 441,963 CpG sites in the validation cohort (Supplementary Table B.1).  41  Chapter 3:!Placental DNA methylation at term reflects maternal serum levels of INHA and FN1, but not PAPPA, early in pregnancy. 3.1! Introduction Early diagnosis of preeclampsia (PE) and intrauterine growth restriction (IUGR) before clinical signs of disease can improve management and outcomes of affected pregnancies. Placental-derived proteins may be released into the maternal circulation where they can be quantified and used to assess placental function during pregnancy (292,293,294,295). Such protein markers have been investigated for the prediction of PE and/or IUGR with varying success (109,113,296). Nicolaides et al. (2013) reported a detection rate of 95% for early-onset PE (EOPE, diagnosis <34 wks) using decreased levels of maternal serum markers, pregnancy associated plasma protein A (PAPPA) and placental growth factor (PLGF), in combination with maternal factors (109). However, these measures might not be generalizable, as the etiology and confounding environmental factors vary between populations (113). Moreover, the ability to predict women at risk of late-onset PE (LOPE, diagnosis ≥34 wks) and IUGR is limited using these markers.   Differential gene expression between placentas from PE and/or IUGR pregnancies (27,147,296,297) may be utilized to identify additional biomarkers to distinguish women at high risk of these complications early in gestation. DNA methylation (DNAm) is associated with gene expression, but is more robust to variation in technical conditions and less subject to short-term biological change (164). Alterations of placental DNAm were noted in genes for which the expression of the encoded protein is altered in maternal blood in PE and/or IUGR pregnancies 42  (e.g.: PAPPA, sENG, PAPPA2) (22). Furthermore, we found that sites of altered DNAm in PE frequently reflected changes in gene expression (22). While proteins produced in the placenta can be released into maternal circulation, their levels in maternal serum may be affected by many additional factors including size of the placenta, the cell type expressing the protein, and how such proteins are transported and metabolized. The purpose of this study was to delineate the relationship between changes we observed in DNAm at term and maternal protein levels in early pregnancy. We selected three genes for which there was evidence for both altered maternal protein levels and altered DNAm in PE (22); we then evaluated 1) the relationship between placental DNAm and gene expression; 2) the role of variables that might confound measurement of DNAm, mRNA or protein levels including gestational age, fetal sex, placental efficiency (fetal: placental weight ratio), birth weight, placental breadth: width ratio and maternal body mass index (BMI); and 3) whether placental DNAm at term reflected protein levels in maternal blood during gestation after correcting for these variables.  3.2! Methods 3.2.1! Sample information We used a total of 171 placentas for our studies, not all placentas were used in all studies as we were limited by samples run on the Illumina HumanMethylation450 array (450K) (N=66); samples run on the Illumina expression array (N=16), maternal serum screening results (first trimester N=34, second trimester N=36), or maternal serum samples for Fibronectin (FN1) testing (N=114).  Supplementary Table C.1 outlines a list of all samples and which analyses they were used in.  43  3.2.2! Gene expression analysis Gene expression was measured with the HT-12v4 Expression BeadChip (Illumina, Inc.) as per Blair et al. (2013) protocol, comparing eight EOPE and eight controls (22) (Supplementary Table C.2).   3.2.3! DNA methylation analysis 3.2.3.1! Illumina Infinium HumanMethylation450 beadchip array To compare the DNAm differences between pathological groups for each of our candidate genes 20 EOPE, 11 LOPE, 8 LOPE+IUGR, 10 IUGR, and 37 control cases were run on the 450K array, which interrogates >480,000 CpG sites in >20,000 genes (271). Some of these samples were previously analyzed in the study reported by Blair et al. (2013).    3.2.3.2! Bisulfite pyrosequencing Candidate CpGs determined from the 450K array data in Blair et al. (2013) were followed up with bisulfite pyrosequencing in control cohorts for each candidate gene (Table 3.1). To compare the association between DNAm and protein levels in maternal blood, 122 placental DNA samples (750ng) were bisulfite converted using the EZ DNA methylation-Gold kit (Zymo Research Corp, Irvine, CA, USA) as per manufacturer’s protocol. Bisulfite converted DNA was PCR amplified prior to pyrosequencing. PCR reactions consisted of 20ng of bisulfite converted DNA, 1x PCR buffer (with MgCl2) (Qiagen Ltd.), 0.18U DNA polymerase (HotStarTaq, Qiagen Ltd.),0.2mM dNTP (Invitrogen, Carlsbed, CA),0.4uM forward and reverse primers (Integrated DNA Technologies, Coralville,IA) for Inhibin beta alpha (INHBA), Pregnancy associated plasma protein A (PAPPA), and Fibronectin (FN1). PCR conditions were 95°C (15min), [95°C (30s), 44  55°C (30s), 72°C (30s)] x 40 cycles, 72°C (10min). Pyrosequencing assays for the candidate genes were designed in PSQ Assay Design software (Biotage, Upsalsa, Sweden) and run on a Qiagen Pyromark Q96 MD (Qiagen) (Supplementary Table C.3).   Table 3.1. Candidate CpG sites chosen from Blair et al. (2013) for follow-up Gene Site Genomic Region Distance to TSS (base pairs) EOPE β value (Δβ#from#control#group) INHBA cg11079619 Active Promoter 76 0.434 (-0.162) PAPPA cg08189448 Active Promoter -163 0.326 (-0.074) FN1 cg12436772 Intergenic/Upstream enhancer -101593 0.465 (-0.240)   3.2.4! Candidate DNA methylation selection CpG sites chosen to investigate in the present study were selected on i) a significant change in placental DNAm, defined as a false discovery rate (FDR)<0.05, and a biological threshold of ∆β>0.05 (i.e. at least 5 percentage points difference in DNAm), a cut-off that enriches for changes in DNAm that would likely have biological impact (298), in placentas associated with PE and ii) genes encoding for proteins reported to show altered levels in maternal blood in pregnancies complicated by PE and/or IUGR. We chose FN1 since the difference in DNAm between EOPE and controls was high (Δβ= 0.24). INHBA (∆β=-0.16, FDR<0.05) and PAPPA (∆β=-0.074, FDR<0.05) were selected because they additionally encode for proteins for which 45  first (PAPPA) or second trimester (INHA) maternal serum measures were available from clinical prenatal serum screening testing.  Previous studies have shown upregulation of both PAPPA and INBHA in the placentas of pregnancies complicated by PE and IUGR (22,297,299,300). In addition, several studies have reported DNAm alterations in placentas from pregnancies complicated by PE and/or IUGR (15,22,301). We also took into account where the DNAm alterations were in the genome, taking interest in alterations in gene regulatory elements (Table 3.1). The CpGs of interest for INHBA and PAPPA were 76 base pairs (bp) and 163bp upstream of the transcriptional start sites, respectively. In relation to FN1, the CpG site was ~100 kb upstream of the transcriptional start site, within an enhancer region.   3.2.5! Maternal blood protein measurements Measurements of PAPPA and INHA were obtained from clinical maternal serum screening data for 36 and 33 women, respectively, and are measured in multiples of the median (MoM).  Blood was drawn in EDTA tubes during the second trimester for a subset of 158 women (Table 3.2). Plasma was obtained via centrifugation at 3000rpm for 10 minutes 4⁰C. Plasma FN1 was measured using a FN1 ELISA kit (eBioscience, San Diego, CA, USA). FN1 measurements were run in duplicate and absorbance was measured at 450nm. A 5 parameter asymmetrical logistic curve was generated from the standard data points which ranged from 0.31-20.0ng/mL. Samples were diluted as per manufacturer’s protocol; samples which FN1 concentration was over the standard curve were diluted to 1 in 80,000, and 4 samples which remained were further diluted to 1 in 100,000.46   Table 3.2. Samples used for pyrosequencing and measuring maternal FN1 protein levels.  Control EOPE LOPE+IUGR LOPE IUGR INHA N= 36 - - - -  Mean GA at blood draw(weeks ±SD) 14-20wks - - - -  Mean GA at delivery (weeks ±SD) 39.3 (±1.3) - - - -  Mean BW (grams ±SD) 3480.3 (±483.4) - - - -  Mean  MA (years ±SD) 33.5 (±4.4) - - - -  Sex (Female/N, %) 18/36, 50% - - - - PAPPA N= 33 - - - -  Mean GA at blood draw(weeks ±SD) 11-13wks - - - -  Mean GA at delivery (weeks ±SD) 39.6 (±1.4) - - - -  Mean BW (grams ±SD) 3428.9 (±355.9) - - - -  Mean  MA (years ±SD) 34.2 (±4.6) - - - -  Sex (Female/N, %) 18/34, 53% - - - - FN1 N= 76 13 6 10 9 47   Control EOPE LOPE+IUGR LOPE IUGR  Mean GA at blood draw(weeks ±SD) 31.6 (±6.1) 32.3 (±3.2) 35.9 (±1.3) 37.4 (±2.4) 33.5 (±4.5)  Mean GA at delivery (weeks ±SD) 39.1 (±2.9) 33.1 (±3.2) 36.1 (±1.1) 38.4 (±1.9) 35.2 (±4.5)  Mean BW (grams ±SD) 3465.3 (±398.94) 1663.3 (±710.5) 1921.7 (±402.5) 3187.0 (±683.3) 1932.2 (±746.0)  Mean  MA (years ±SD) 33.5 (±3.6) 33.4 (±6.4) 32.4 (±5.3) 35.5 (5.5) 33.5 (±3.5)  Sex (Female/N, %) *36/74, 49% 6/13,46% 3/6, 50% 6/10, 60% 6/9,66% *Sex%not%available%on%2%samples%48  3.2.6! Statistical analysis DNAm at the two CpGs in the PAPPA pyrosequencing assay were correlated (r=0.85, p<0.001, Spearman’s correlation) and the measurements for these two sites were thus averaged (Supplementary Figure C.1). Potential covariates which may be associated with either DNAm or protein concentration in maternal blood were assessed for each candidate site. Univariate linear regression analyses were performed, investigating gestational age at delivery, fetal sex, birth weight (SD), fetal: placental weight ratio, placental length: breadth ratio, maternal body mass index (BMI), and when appropriate, gestational age at blood draw. As absolute birth weight is confounded by gestational age at delivery, birth weight was measured as a standard deviation relative to the mean for that gestational age and infant sex. PAPPA and INHA protein levels were expressed in MoM to correct for gestational age at blood draw.   Correlations were performed when testing any association between placental gene expression at term and placental DNAm at term. Spearman’s correlations were performed between protein concentration and DNAm in sites where there were no covariate factors. For sites with covariate factors, partial correlations were performed. Non-parametric t-tests were performed to determine if DNAm in the EOPE, LOPE+IUGR, LOPE, and IUGR placentas were significantly different from controls. Statistics were calculated using SPSS v19.0 statistical package.  3.3! Results 3.3.1! Candidate site selection and characteristics Although candidate sites were selected based on a significant association with EOPE (See methods) (22), we also wanted to know if these changes were conserved in other pathological 49  groups (Figure 3.1). In addition to hypomethylation of these sites in EOPE, the LOPE+IUGR group was hypomethylated for the INHBA (promoter) (∆β=-0.18, p<0.001) (Figure 3.1a) and the FN1 upstream enhancer (∆β=-0.25, p<0.01) (Figure 3.1c).  While reduced DNAm at the PAPPA promoter was only found in EOPE (Figure 3.1b). 50   Figure 3.1. DNA methylation measures at candidate sites.  β values ±SD obtained from microarray data (450K) at each site across pathological groups for A) INHBA, B) PAPPA, and C) FN1.  EOPE= early-onset PE (N=20), LOPE=late-onset PE (N=11), IUGR=Intrauterine growth restriction (N=12), Control (N=37). *p<0.0551  3.3.2! Is DNA methylation at candidate sites inversely correlated with gene expression? To confirm that the DNAm change resulted in a change in gene expression, we assessed the relationship between placental DNAm (by 450K) and gene expression at these three candidate sites. FN1 showed an inverse correlation between DNAm of an upstream enhancer and gene expression at term (r=-0.88, p<0.0001). INHBA and PAPPA, showed a non-significant trend with increasing DNAm being associated with decreased gene expression in the placenta (Figure 3.2). For all candidate genes, there was an observable divide between the controls and EOPE cases, where cases had decreased DNAm corresponding to increased gene expression in the placenta.52   Figure 3.2.  Correlation between DNA methylation and left) gene expression, right) protein levels.  Correlation between DNA methylation and gene expression in eight early-onset PE and eight control placentae in A) INHBA B) PAPPA and C) FN1 †FN1 graph produced from data published in Blair et al. (2013). Relationship between D) INHBA (N=36) promoter DNAm in the term placenta and second trimester INHA levels in maternal blood, plotted as residuals corrected for fetal birth weight (SD) and fetal: placental ratio, E) PAPPA (N=34) promoter DNAm in the term placenta and first trimester PAPPA levels in maternal blood, and F) FN1(N=76) enhancer DNAm in the term placenta and second/third trimester FN1 levels in maternal blood, plotted as residuals corrected for fetal birth weight (SD), gestational age, and maternal body mass index (BMI). MoM=multiple of the median. 53  3.3.3! What clinical factors are associated with DNA methylation at candidate sites? To better understand what factors might affect the measurement of DNAm and therefore the relationship with protein expression levels in maternal blood, we also evaluated several potential confounding factors including gestational age at delivery (277), fetal sex (302,303), fetal birth weight (304), placental dimensions and maternal BMI. Bisulfite pyrosequencing was used to extend our assessment of DNAm at the candidate sites into a larger cohort of controls for which clinical serum measurements (INHBA N=36, PAPPA N=33) or serum samples for assaying FN1 (N=76) were available.     Birth weight standard deviation (SD) was associated with DNAm at the INHBA promoter (p=0.05) and the upstream enhancer of FN1 (p=0.02). Gestational age was only associated with FN1 DNAm (p=0.03). None of the clinical factors assessed was associated with DNAm at the PAPPA site (Table 3.3).  Table 3.3. Univariate linear analysis results for DNA methylation vs. clinical parameters.  Samples are all controls. Reported in correlation coefficient (r) values. *p<0.05 Gene N= Fetal Sex Gestational Age at Delivery Birth Weight (SD) Fetal: Placental Weight Placental Length: Breadth Maternal BMI (Number of samples BMI was available) INHBA 36 0.53 0.055 0.29* 0.08 0.01 0.35 (N=18) PAPPA  34 0.18 0.17 0.24 0.00 0.26 0.25 (N=21) FN1 76 0.12 0.22* 0.23* 0.10 0.23 0.30 (N=75, all samples)        0.12 (N=37,control only) 54  3.3.4! What clinical factors are associated with protein concentration in maternal blood? We also assessed the same clinical parameters for association with protein concentration in maternal blood (Table 3.2). Gestational age at blood draw was only assessed as a covariate for FN1 as clinical values for INHA and PAPPA were given in multiples of the median (MoM), which was already corrected for GA at blood draw. Placental efficiency (fetal: placental weight-ratio, at birth) was associated with increased second trimester INHA levels in maternal blood. FN1 level was not associated with maternal BMI in the controls for which we had this information (N=37), though it was significant when evaluating all clinical groups together (EOPE, LOPE, IUGR, Controls) (N=75). It was therefore included in subsequent analyses. None of the assessed factors were associated with PAPPA maternal blood levels during pregnancy (Table 3.4).   Table 3.4. Univariate linear analysis results for protein Levels vs. clinical parameters. Samples are all controls. Reported in correlation coefficient. Protein N= Fetal Sex Gestational Age at Delivery Gestational Age at Blood Draw† Birth Weight (SD) Fetal Weight:Placental Weight Placental Length:Breadth Maternal BMI (Number of samples BMI was available)  INHA 36 0.20 0.00 NA 0.12 0.44* 0.30 0.34 (N=18)  PAPPA 34 0.11 0.20 NA 0.26 0.08 0.00 0.05 (N=21)  FN1 76 0.05 0.10 0.16 0.11 0.13 0.063 0.25* (N=75, all samples)          0.10 (N=37, control only)  *p<0.05'†'Only'measured'for'FN1'as'INHA'and'PAPPA'levels'were'obtained'from'maternal'serum'screening'program'and'already'corrected'for'gestational'age'at'blood'draw.!  55  3.3.5! What is the relationship between DNAm and maternal serum levels? DNAm in the promoter of INHBA correlated with second trimester protein levels in maternal blood (r=-0.50, p=0.01) while modeling for both birth weight (SD) and fetal: placental weight ratio (Figure 3.2d). Additionally, DNAm in an upstream enhancer of FN1 correlated with third trimester protein levels in maternal blood (r=-0.38, p=0.009) while adjusting for birth weight (SD), gestational age, and maternal BMI (Figure 3.2f). In contrast, a similar result was not observed for PAPPA/PAPPA (Figure 3.2e).   3.3.6! Are there any differences in protein levels between case and control placentas? To confirm a previous report of altered maternal FN1 in association with PE and/or IUGR(305), FN1 levels were measured in maternal blood samples from pregnancies which subsequently developed EOPE, LOPE+IUGR, LOPE without IUGR, or normotensive IUGR, in addition to our control cohort (Table 3.2). Similar to the alterations in DNAm, changes in FN1 levels were found to be significantly different from controls only in the EOPE group (Mann U Whitney test), although there was a trend of increased FN1 levels between LOPE+IUGR and controls (p=0.08) (Figure 3.3). 56  Control EOPE LOPE+IUGR LOPE IUGR050010001500** *FN16(ug/mL) Figure 3.3. Fibronectin 1 (FN1) levels in each pathological group. FN1 levels (Median with interquartile range) in maternal blood during third trimester. EOPE= early-onset PE (N=20), LOPE=late-onset PE (N=11), IUGR=Intrauterine growth restriction (N=12), Control (N=37). **p<0.05, *p<0.1  3.4! Discussion The differences in DNAm in our candidate genes were only found in the EOPE and LOPE+IUGR groups. Therefore the utility of these potential candidate genes/ biomarkers would be limited to identifying this subset of pregnancies (306). Markers useful to detect LOPE or normotensive IUGR may be more challenging to identify due to their weak association with placental pathology.   The observed relationship between DNAm in the term placenta to second trimester (INHBA/INHA) and third trimester (FN1/FN1) maternal serum levels supported our prediction that DNAm changes observed in the placenta could explain some of the previous reports of altered INHA and FN1 levels in maternal blood in PE. It is remarkable that these serum 57  measurements from the second and third trimesters of pregnancy reflected DNAm at term. This implies that this DNAm change may be an early alteration in PE.  We had predicted that protein levels in maternal blood would reflect placental DNAm and gene expression. While this may be true in some instances (e.g. INHA, FN1), in other cases establishing a relationship may be challenging (e.g. PAPPA/PAPPA). Establishing such a relationship may be complicated by several factors. Protein level depends not only on the level of gene expression, but also on the total number of cells expressing that protein, the number of mRNA transcripts being translated into protein in those cells, and the rate and mode of release of the protein into maternal blood. These factors may be influenced by the underlying pathology (i.e. more protein may be released with increased apoptosis) and placental size; which, in turn may be associated with fetal weight and/or fetal: placental weight ratio. Other factors such as expression of the same protein from maternal tissues, and the metabolism of proteins by the placenta, reducing the amount of protein being secreted into the maternal circulation, may have a substantial influence of the total protein concentration in maternal blood (Figure 3.4). PAPPA has been found to be expressed from other maternal sources (e.g. ovary, some epithelial and endometrial cells, and breast) besides the placenta, and it is possible that these sources mask any relationship between placental derived protein and DNAm in the placenta (307,308,309,310). It is also important to note that PAPPA maternal protein levels were measured in the first trimester and additional variation may arise over gestation affecting correlation with placental DNAm at term.58   Figure 3.4. Processes that may influence the relationship between DNA methylation, gene expression and protein expression.  Outlines reasons why we may not see a correlation between placental DNAm and gene expression or between placental gene expression and circulating levels of placental-specific proteins in maternal blood.59  !Our results were in concordance to Auer et al. (2010) who also reported increased levels of maternal FN1 in pregnancies complicated by EOPE and LOPE+IUGR (305). We did not confirm their observation of a decrease of FN1 in pregnancies complicated by IUGR; however, we may have been under-powered to observe this small difference. Furthermore, although we observe a difference in EOPE and LOPE+IUGR compared to controls, the range of FN1 levels completely overlap between the groups, hindering FN1 to be an adequate biomarker when used alone.  This study provides a link between changes in placental DNAm at term and protein biomarkers present in the mother’s circulation earlier in pregnancy. It emphasizes the many confounding factors that may influence this relationship, explaining why this linkage may not be observed for all loci. We chose three genomic sites with significantly altered DNAm in term placenta associated with PE and that were associated with genes for which the protein product is altered in PE/IUGR. We found a correlation between placental DNAm of INHBA and FN1 at delivery and second and third trimester maternal serum protein expression, respectively. This may suggest that other DNAm marks may be associated with early differences in gene expression. Furthermore, with the advent of techniques to quantify placental nucleic acids in maternal serum (114), DNAm changes may be more directly linked to measurable miRNA and RNA in maternal blood. Factors such as placental surface area and mechanisms for release into maternal blood, will also affect serum levels of placental nucleic acids (311). Future studies measuring protein levels directly in placental tissue, correlating with maternal levels and investigating the factors affecting rate of release are needed to help translate findings measured in the term placenta into maternal biomarkers of pregnancy outcomes in early gestation. Following up DNAm alterations 60  in future studies, it is imperative to assess DNAm alterations that are reproducible. For this reason, future studies should validate their finding in an independent cohort.  61  Chapter 4:!Utility and validation in placental DNA methylation profiling of preeclampsia and intrauterine growth restriction 4.1! Introduction Both preeclampsia (PE) and intrauterine growth restriction (IUGR) are heterogeneous in etiology, with many different factors contributing to these phenotypes (67,280,312). Due this heterogeneity, the ability to sub-classify placentas into more homogeneous groups can aid in our understanding of disease pathogenesis and prediction. For example, by defining ‘placental IUGR’ on the basis of a detailed scoring system for placental pathology, Benton et al. showed this subset was associated with very low maternal serum placental growth factor (PLGF) and also had the most severe perinatal and postnatal risks (313). In some cases, PE and normotensive IUGR (nIUGR) may represent two facets of a common underlying etiology, while in others the associated placental pathology and molecular changes may be distinct. Enforcing stringent criteria for defining and grouping samples may increase the reproducibility for reported molecular changes. For this study, we subdivide our samples into early-onset PE (EOPE), late-onset PE (LOPE), and nIUGR based on clinical obstetric criteria.   Molecular profiling has the potential to refine these clinically-defined group definitions further and improve our current understanding of the etiology of EOPE, LOPE, and nIUGR and their relationship to one another. Placental transcriptome profiling from pregnancies associated with PE and healthy controls provide evidence for multiple subtypes of PE (20). DNA methylation (DNAm) profiling is an alternative or complementary approach to gene expression profiling to identify subgroups of placental phenotypes. DNAm is more stable than mRNA and hence is less 62  subject to changes with sample processing time (164); it may also retain a “memory” of earlier in utero exposures and hence be linked to early effects in the disease process.    The aims of the present study were to build on our understanding of DNAm changes in placental insufficiency and to i) reanalyze our previous EOPE data with more sophisticated approaches; ii) extend our analysis to LOPE and nIUGR to investigate potential relationships between the three conditions and, iii) validate differentially methylated sites in an independent cohort. We will also discuss challenges to validation and future directions for epigenetics in the placental biology field.  4.2! Methods 4.2.1! Sample information 4.2.1.1! Discovery cohort The discovery (Vancouver) cohort consisted of 22 EOPE, 18 LOPE, 11 nIUGR and 43 control placentas (Table 4.1). A subset of 18 EOPE samples and 19 preterm control samples in this study was previously used in Blair et al. (2013) (22). 63  Table 4.1. Sample information on the discovery and validation cohorts. Discovery!Cohort!! EOPE!!(mean)!LOPE!!(mean)!IUGR!!(mean)!Preterm!“Control”!(mean)!Term!Control!(mean)!N=! 22! 18! 11! 24! 19!Fetal!birth!weightJ!Standard!deviation!range!J8.18!–!3.77!!(J1.65)1!J2.9!–!2.57!(J0.96)2!J2.57!J!J1.22!!(J1.99)3!J1.61!–!3.23!(0.51)!J0.94J!0.98!(J0.09)!Fetal!Sex!(M:F)!14:8! 10:8! 4:7! 16:8! 9:10!Maternal!Age!(years)!19.7J42.9!!(33.3)!23.1J41.3!(34.0)!33.3J38.0!(34.3)!22.2J41.1!(32.5)!30.0J40.2!(34.9)!Gestational!Age!(weeks)!24.9J38.4!(32.0)!34.6J41.3!(37.4)!34.6J38.0!(36.6)4!25.0J36.7!(32.6)!37.3J39.9!(38.4)!Validation!Cohort!! EOPE!(mean)!LOPE!!(mean)!IUGR!!(mean)!Preterm!“Control”!(mean)!Term!Control!(mean)!N=! 22! 11! 0! 6! 9!Fetal!birth!weightJ!Standard!deviation!range!J2.46J!0.01!(J1.40)5!J2.92J!0.01!(J0.84)6!NA! J1.02J!0.05!(J0.16)7!J1.11J!3.60!(0.61)!Fetal!Sex!(M:F)!9:13! 7:4! NA! 4:2! 6:3!Gestational!Age!(weeks)!26.0J34.0!(30.5)!35.0J37.0!(36.4)8!NA! 27.0J33.0!(30.8)!38.0J40.0!(38.8)!Discovery!Cohort:!1!p=!6.9eJ6!vs!PTB!“Control”,!2!p=0.014!vs!Term!Control,!3!p=!1.7eJ5!vs!Term!Control,!4!p=3.3eJ4!vs!Term!Control!Secondary!Cohort:!5!p=0.004!vs!PTB!“Control”,!6!p=0.0005!vs!Term!Control,!8!p=0.0001!vs!Term!Control!!Between!Cohorts:!7!p=0.04!PTB!(discovery)!vs!PTB!(secondary)!64  Technical batch effects, related to the plate, microarray chip, and sample position on the Illumina Infinium HumanMethylation450 (450K) are potential confounding factors within our data. Our samples were run in five batches over a 4 year period, and pathology and gestational age were confounded with batch as EOPE and preterm controls were largely run earlier. In this situation, correction for batch effects can introduce spurious findings (290) (Supplementary Figure D.1 and Supplementary Figure D.2). We, therefore, instead compared EOPE to preterm birth controls and LOPE/nIUGR to term controls only, which were relatively matched for batch, and thus the confounding by gestational age and its interaction with batch was minimized. We acknowledge that some of the differentially methylated sites that we found may be due to technical artifacts, but focusing on those hits that are reproduced in the validation cohort largely eliminated these effects.   As placental DNAm changes with gestational age, the comparison groups included placentas from healthy term births (≥37 weeks) and preterm births (<37wks) with normally grown babies and no evidence of maternal hypertension. EOPE placentas were compared to 24 preterm controls (as in Blair et al. 2013). LOPE and nIUGR placentas were compared to a separate set of 19 term control placentas. This was to test for overlap between DNAm changes identified for LOPE and nIUGR with those for EOPE. We did not want the use of a common control group driving any potential overlap. To reduce the chance of differences being driven by the preterm birth group, we used placentas from preterm births from a variety of etiologies (e.g. premature rupture of the membranes, incompetent cervix, chorioamnionitis), while any term control samples with evidence of pathology involving the chorionic villi were excluded.  65  4.2.1.2! Validation cohort The validation (Toronto) cohort consisted of 22 EOPE, 11 LOPE, and 15 control placentas (Table 4.1). For the validation cohort, placental samples were purchased through the Research Centre for Women’s and Infants’ Health BioBank (Mount Sinai Hospital), details in the sample processing can be found in Leavey et al. (2016) (21). DNA was extracted from the pooled placental tissue by ethanol precipitation using the Wizzard® Genomic DNA Purification Kit (Promega). Gestational age at delivery and fetal birth weight were collected for each case. For this cohort, ethics approval was obtained from both Mount Sinai Hospital (#13-0211-E) and the University of Toronto (#29435).   The validation cohort represents a subset of samples from the Leavey et al. (2016) study (21). PE was defined as BP > 140/90mm Hg after 20 weeks gestation and proteinuria >300mg/day or >2+ by dipstick (314). As the time of diagnosis was unknown, we subdivided the PE samples from this cohort into EOPE and LOPE based on the gestational age at delivery. Exclusion criteria included diabetes, sickle cell anemia, morbid obesity, and multi-fetal pregnancies. The division of the term and preterm controls was also done in the validation cohort, which consisted of 6 preterm control and 9 term control placentas.  !4.2.1.3! DNA methylation analysis Samples were run on the 450K array. To minimize any effects of sample processing between the validation and discovery cohorts, arrays were run in the same batch and with the same operators as a subset of the samples from the discovery cohort (Chips 5013, 5015, 3024,3037,3038,3110, See Supplementary Figure D.1 and Supplementary Figure D.2). This DNAm data for the 66  discovery and validation cohorts is available from the Gene Expression Omnibus (GEO) database under the accession numbers GSE100197 and GSE98224, respectively. !4.2.1.4! Differential DNA methylation analysis All statistical analyses were performed using R version 3.2.4. Differentially methylated sites were identified using statistical, i.e. false discovery rate (FDR) <0.05, and biological, i.e. a change in DNAm (Δβ)>0.1, criteria. We corrected for fetal sex in our linear regression model, but we did not adjust for birth weight, as it is closely related to pathology. As our groups were matched to controls of a similar gestational age, our final model where DNAm alterations were identified took into account fetal sex only. Only those sites that met both these criteria were then evaluated in the validation cohort. In this case, linear regression was used, and sites were considered to be persistent hits if the nominal p-value <0.05 and the change in DNAm was in the same direction as the discovery cohort. Bonferroni correction p <0.05 was also used on the 1703 EOPE differentially methylated sites to investigate how many hits would be validated with a more stringent threshold. To compare the heterogeneity of the samples in each cohort, sample by sample Pearson’s correlations (correlating all DNAm measures in each sample to every other sample in the study) were performed, and the average correlation of each sample was compared between the two cohorts by Student’s t-test separately for the control (Term + Preterm) and EOPE samples.  To investigate whether the Bonferroni corrected hits and the nominal p-value validated hits were more than would be expected by chance, 1703 sites (number of EOPE hits in the discovery cohort) were randomly sampled from the validation cohort data and run through a linear model, 67  correcting for fetal sex. One thousand permutations were run and the number of sites that met a nominal p-value<0.05 in each iteration was recorded. The number of randomly sampled sites to meet a nominal p-value<0.05 were compared to the actual number of sites that validated in our data (N=42 (Bonferroni corrected) and N=599 (nominal p-value)).   4.2.1.5! Clustering analysis Hierarchical clustering was performed on the persistent hits to investigate whether samples clustered according to their clinically diagnosed pathology, or whether DNAm profiling could suggest an improved definition of pathological groups. The pvClust package in R (315) assessed how stable any resulting clusters were, using 1000 iterations. The sigClust2 package (316) determined if any clusters were significantly different from one another, also using 1000 iterations. To investigate whether differences in DNAm between the clusters were enriched for any specific pathway(s), linear regression was used to identify differentially methylated sites between clusters. Differentially methylated sites were annotated to genes using the Price et al. annotation, closest transcriptional start site (291), and then input into ermineJ, a gene ontology tool (317). ErmineJ allows us to input a background gene list specific to the Illumina 450K array, accounts for multifunctionality (gene ontology terms that appear frequently due to the number of genes involved in the pathway), and allows for multiple iterations to be run to strengthen the power of the analysis.   68  4.3! Results 4.3.1! Widespread DNAm changes are associated with EOPE but not LOPE and nIUGR in our Discovery cohort Our first goal was to confirm our previous report of widespread changes in EOPE (22) and then to test for similar changes in LOPE and nIUGR using the same approach. We chose less stringent cutoffs for significance in this analysis (FDR<0.05 & Δβ>0.1) as compared to Blair et al. (2013) (FDR<0.01 & Δβ>0.125) (22) as our aim was to identify a larger number of differentially methylated sites that could be used for further validation. Based on these criteria, a total of 1703 sites were differentially methylated between EOPE and preterm controls (Figure 4.1). As expected, the majority (261/286) of EOPE hits reported in Blair et al. were also identified as hits in this analysis. Differences between the two analyses are likely explained by the use of different normalization methods, correction for fetal sex in the present study, and the inclusion of a few additional samples in this study compared to the previous one.   We used the same approach to identify differential methylation associated with LOPE or nIUGR as compared to the healthy term control group. In contrast to the EOPE comparison, only 5 sites were differentially methylated between LOPE and term controls, and no sites were differentially methylated between nIUGR and term controls (Figure 4.1). The 5 differentially methylated sites between LOPE and term controls were not unique to LOPE, as they were also included amongst the 1703 sites identified as differentially methylated in EOPE. The weaker signal may be explained if only a few of the LOPE cases have an underlying pathology similar to that in the EOPE cases, driving these changes. 69  !Figure 4.1. Volcano plots depicting differentially methylated sites.  Differentially methylated sites between A) EOPE and preterm controls (preterm-EOPE), B) LOPE and term controls (term-LOPE), and C) IUGR and term controls (term-IUGR). –log10 of the adjusted p-value is plotted on the y axis and the change in DNAm (∆β) is plotted on the x axis. Sites highlighted in red are hypermethylated in the pathology compared to controls. Sites highlighted in green are those that are hypomethylated in the pathology compared to controls.70  4.3.2! Validation of the EOPE hits in an independent cohort  We next investigated if the EOPE hits from our discovery cohort could be validated in an independent cohort.  We first tested whether the ∆β values in the discovery and validation cohorts were correlated using all sites that met an FDR<0.05 in the discovery cohort, without imposing an additional ∆β threshold. At these sites, the correlation was significant (R=0.62, p<2.2e-16, Figure 4.2a). This indicates that largely similar changes in DNAm are being observed in the EOPE placentas in both cohorts. Amongst the most highly significant hypomethylated sites (∆β >0.15) in both cohorts were CpGs associated with KRT15, FN1, TEAD3, JUNB, ST3GAL1, PKM2, NDRG1, PAPPA2, CHI3L2, and INHBA. Many of these genes have previously been shown to have altered gene expression in preeclampsia and JUNB has been specifically implicated as a key player in the response to hypoxia in trophoblast cells (318). Amongst the most highly hypermethylated sites (∆β <-0.10) in both cohorts were several sites associated with FAM3B, SYNE1, and AGAP1. However, it should be noted that there were also many sites with a high Δβ in the discovery cohort that had a much smaller or sometimes opposite direction Δβ in the validation cohort.   To narrow down the original 1703 EOPE hits from the discovery cohort to a high-confidence hit list, we first asked, how many of these hits met similarly stringent criteria (FDR<0.05 and ∆β>0.1) in the validation cohort? We found that only 38 probes (2.2%) met these strict criteria.  Using such arbitrary cut-offs in both populations and a strict definition for a “hit” may not be a powerful approach to assess the degree of overlap in the data. Furthermore, requiring assay-wide correction for multiple testing in the validation cohort is overly conservative and reduces power. Running the linear regression on only the 1703 sites differentially methylated in the discovery 71  cohort reduces the number of multiple test corrections needed in the validation cohort. Based on the distribution of nominal p-values among the 1703 EOPE associated sites in the validation cohort, shown in Figure 4.2b, there are many more sites that meet a nominal p-value<0.05 than expected, even if these do not meet a multiple test correction. Hence, we opted to use a nominal p-value<0.05 and a change in DNAm in the same direction as the discovery cohort to define validated (i.e. high confidence) hits. Based on these criteria, 599 of the 1703 (35.1%) EOPE hits were considered to be validated (Figure 4.2b). This is higher than what we would expect by chance (p=0.0001). This reproducibility rate was similar to the rate reported by Yeung et al. (2016), who validated their own differentially methylated regions with our published cohort (Blair et al. (2013))(240). As PE and IUGR are heterogeneous conditions, it is possible that the reproducibility rate may be affected by the samples chosen for each cohort. We were interested in whether the samples in the cohorts were similarly correlated (i.e. is one cohort more heterogeneous than the other).  We investigated these correlations in the control samples (term and preterm) (Supplementary Figure D.3) and the EOPE samples (Supplementary Figure D.4). In both pathologies, samples in the discovery cohort were more heterogeneous (more variability in DNAm measures) than the samples in the validation cohort. 72  !Figure 4.2. Validation of DNA methylation alterations. A) The correlation between the change in DNA methylation (∆β values) between EOPE and preterm controls (preterm-EOPE), between the discovery and validation cohorts (R=0.62, p<2.2e-16). The sites highlighted in red are the top sites labeled by the gene the CpG site is located in. B) P-value distribution of the 1703 EOPE hits from the discovery cohort, in the validation cohort. There were 599 sites of the 1703 EOPE sites (35.1%) that validated.73  These validated sites were not enriched for any gene ontology terms using ermineJ, with a 450K array specific background (317). A list of these sites and relevant gene information can be found in Supplementary Table D.1. These sites include ones associated with genes known to be relevant to EOPE from gene expression studies including CGA, INHBA, PAPPA2, and ADAM12.  4.3.3! Effects of varying validation criteria To evaluate the effect of varying FDR and Δβ cutoffs to establish the most ‘reproducible’ results, we plotted the percentage of probes that showed Δβ concordance in directionality between the validation cohort and the discovery cohort using different FDRs and Δβ thresholds in the discovery set (Supplementary Figure D.5a). Different FDR thresholds did not influence DNAm concordance rate when the ∆β thresholds were above 0.2. FDR thresholds appear to be more important when trying to identify small changes in DNAm.  We also investigated the number of hits that each threshold would obtain. Supplementary Figure D.5b plots the number of hits at each FDR and Δβ cutoff. Allowing smaller changes in DNAm produces many more hits, but with a lower reproducibility rate. This is likely because this is in the range of normal variability for a site. This highlights the importance of considering both the biological and statistical thresholds depending on the magnitude of the anticipated DNAm change and the overall research objective.   4.3.4! Hierarchical Clustering Next, we evaluated the degree to which the 599 validated sites can discriminate EOPE from all other placentas in both cohorts, using hierarchical clustering (Figure 4.3). Although we expected an EOPE methylation cluster to be defined, since we are clustering based on our EOPE hits, this 74  approach can inform us about the relationships between individual samples and, furthermore, allow comparisons to term controls, and LOPE and nIUGR cases which were not involved in the selection of these EOPE hits. Interestingly, both cohorts clustered into 3 stable methylation clusters, rather than just two as we had expected. When the cohorts were clustered on the 599 validated hits separately, methylation cluster 1 included almost all EOPE suggesting a consistent phenotype in this group. In the discovery cohort cluster, 6 LOPE samples clustered with the EOPE samples. In the validation cohort, 7 LOPE samples and 1 preterm control clustered with the EOPE samples. Additionally, in both cohorts, methylation sub-clusters were identified within the larger EOPE group (methylation cluster 1), which were also stable and significantly different from one another, suggesting a possible further subdivision or distinct groups within methylation cluster 1. There was no obvious difference between these subclusters clinically (including sex, ethnicity, disease severity etc.); however, gestational age was decreased in the EOPE methylation subcluster 1 compared to subcluster 2 (p<0.01) in the validation cohort (Table 4.2)75  !Figure 4.3. Hierarchical clustering (Euclidean) on the 599 validated hits in both the discovery (left) and validation (right) cohorts.  Numbers represent the percentage of times these clusters formed when using 1000 iterations with pvclust. Those highlighted in green are considered stable, where clusters formed >75% of the time. P values signify clusters are significantly different from one another.76  Table 4.2. Clinical information on samples assigned to methylation clusters.  Methylation cluster 2 was compared to methylation cluster 3 and EOPE methylation subcluster 1was compared to EOPE methylation subcluster 2.  77  The remaining non-EOPE samples also clustered into two methylation clusters in both cohorts. We refer to these clusters as methylation clusters 2 and 3. Methylation cluster 3 in both cohorts was predominantly composed of controls. Within the discovery cohort, methylation cluster 2 consisted of the majority of the nIUGR and LOPE cases, a few EOPE cases, and some preterm and term controls; Decreased birthweight (p<0.01) and a trend towards increased IUGR diagnosis (p<0.1) was observed in methylation cluster 2 vs. 3.  In the validation cohort, methylation cluster 2 consisted of preterm controls and LOPE samples and was associated with decreased gestational age (p<0.01) (Table 4.2).   4.3.5! Cluster gene ontology As it was unexpected that the control samples divided into two distinct methylation clusters, and the EOPE samples divided into two distinct subclusters, we were interested in investigating the differences between the two control methylation clusters (methylation clusters 2 and 3) and between the EOPE methylation subclusters (EOPE methylation subclusters 1 and 2). Linear regression was used on the 599 persistent hits, accounting for fetal sex, to identify DNAm differences between methylation cluster 3 and methylation cluster 2 and between EOPE methylation subcluster 1 and EOPE methylation subcluster 2. There was no enrichment in gene ontology terms in either ermineJ (with a 450K array specific background), or DAVID. Between methylation cluster 2 and methylation cluster 3, 244 sites were differentially methylated in both cohorts. Information on these sites can be found in Supplementary Table D.2. There was no gene ontology enrichment by ermine or DAVID. Between EOPE methylation subcluster 1 and EOPE methylation subcluster 2, 207 sites were differentially methylated in both cohorts. Information on these sites can be found in Supplementary Table D.3. There was no gene ontology enrichment by 78  ermineJ and symporter activity was the only gene ontology term in DAVID to meet multiple test corrections.  !4.4! Discussion We previously reported widespread changes in DNAm associated with EOPE (22). In the present study, we extend this analysis to LOPE and nIUGR; however, using the same approach, we were unable to identify DNAm changes that were unique to these groups. While these latter comparisons were limited by small sample size, we were able to obtain significant associations with EOPE with similarly small sample sizes. The reduced number of changes in the LOPE and nIUGR groups can occur for two main reasons: 1) there may be much more limited placental pathology with these diagnoses and the phenotype is largely driven by maternal factors, or 2) they may be more heterogeneous etiology thereby limiting power to detect changes in the group as a whole. If we want to improve biomarker discovery in these groups, we may need to identify more homogeneous subgroups using a combination of clinical parameters, pathology reports and/or already proposed biomarkers, along with larger sample sizes.   The LOPE samples in the discovery cohort that clustered with the EOPE samples all presented with PE between 34.0 weeks and 35.9 weeks gestation and had co-occurring IUGR. While it’s possible that PE symptoms were present but not diagnosed until after 34.0 weeks, there may also be inaccuracies in dating the pregnancy and/or there is simply a grey zone in the distinction between EOPE (placenta-driven) and LOPE (maternal-health driven). It is important to note that the 34 week cut-off between EOPE and LOPE was chosen somewhat arbitrarily to note that prematurity has a much more severe effect on the baby. Therefore, not all cases likely influenced 79  by ‘placental-mediated’ PE will be correctly classified by the 34 week definition of EOPE and LOPE. There were also four cases of EOPE within the discovery cohort that did not cluster with other EOPE cases. One was diagnosed with hemolysis elevated liver enzymes and low platelet (HELLP) syndrome and delivered at 33.3 weeks gestation, one also had chorioamnionitis (which may have contributed to early delivery); one had preexisting hypertension and was diagnosed early but did not deliver until 37 weeks and hence may have been milder in presentation; the fourth was delivered at 33.3 weeks with no other placenta or maternal health notes. Of note, none of the EOPE cases that clustered outside of the EOPE cluster had co-occurring IUGR and were generally diagnosed at close to 34 weeks gestation. Thus, the presence of IUGR in cases diagnosed between 32 and 36 weeks may be the more defining feature as to whether an altered placental DNAm profile, and therefore placental-mediated disease is observed or not. Powers et al. (2012) showed that there are two types of PE pregnancies: those with and without altered angiogenic factors (26). As alterations in the angiogenic factors have also been observed in IUGR cases (319,320), Myatt and Roberts suggested that an imbalance in these factors may represent a measure of placenta growth, development, and function (321). As such, the EOPE cases clustering outside the EOPE methylation cluster may be more likely related to other contributing factors than placental dysfunction.  While we expected to see an EOPE methylation cluster, as the validated hits were chosen based on differentially methylated sites between EOPE and preterm controls, we were surprised that both the EOPE and control groups each formed two subclusters. The driving differences between these subclusters were not clear, though the tendency to lower gestational ages and fetal birth weights (SD) in cluster 2 could suggest features linked to preterm birth (Table 4.2). The presence 80  of two distinct subclusters within the EOPE methylation cluster could reflect PE severity, or perhaps unmeasured factors such as medical treatments given or duration of hypertension. Unfortunately, we had insufficient information on the treatment of each case to evaluate the influence of medical care on the placental methylation profile.  While altered placental DNAm has been reported for pregnancies complicated by PE (13,16,246,322) and IUGR (50,53,323), only one study validated their findings, using the same technology, in an independent cohort with 34.7% of their differentially methylated regions validating (240). In this study, 35.1% (N=599) sites were found to be differentially methylated between EOPE and preterm controls in both the discovery and validation cohorts using validation criteria of a nominal p<0.05 and change in DNAm in the same direction as the discovery cohort. The extent of validation, however, is dependent on the initial criteria chosen to define ‘hits’, the criteria for validation, the similarity of the populations of samples, the size of the study populations (power to detect changes), and the similarity in processing the samples. In our study, the validation cohort was from a roughly similar urban population (Vancouver vs. Toronto) from the same country (Canada). We also tried to minimize technical factors that may influence results by using similar placental sampling protocols, processing the arrays with a subset of the discovery and validation cohorts on the same microarray chips at the same time, with the same technicians, and using the same pre-processing methods on the raw data. Even with these considerations, a significant number of our original hits were not validated. This may be because of chance variation in causes of PE and IUGR in the two cohorts due to limited sample size. Additionally, there are genetic, environmental, and maternal factors that pre-dispose a pregnancy to developing placental insufficiency, which may have varied between populations.  81   Changes in DNAm could mean i) an average change in DNAm across our sample, or ii) a change in the cell type proportions within a sample, as DNAm varies widely across different cell types (210). In the context of EOPE, DNAm alteration may reflect a combination of altered gene expression pathways associated with PE (e.g. related to known effects such as oxidative stress and altered angiogenesis (26,67)) or altered cell type proportions related to PE pathology (e.g. decreased proliferation of extravillous trophoblast cells or alterations to the rate of trophoblast proliferation (324)).  As cell-type specific profiles have not been developed for all placental cell types, it is not possible to use the DNAm profile to estimate cell proportions, as it has been applied to blood (213). While reference-free methods for deconvolution of cell proportions have been developed (325), these methods remove variance within the data attributed to cell composition but cannot inform us of what cell types specifically are altered in EOPE.  Our data demonstrate some of the challenges in identifying changes specific to clinically defined etiologies. Heterogeneity and milder phenotypes in LOPE and nIUGR likely limit the power to detect differences using a differential methylation type approach and mask the subset of cases that do exhibit altered pathology (based on sample clustering using our EOPE defined hits). An alternative approach may be to reduce the dimensions in the data by 1) removing non-variable probes across all cell types (326), 2) focusing on alterations in pathway modules, as in weighted gene co-expression network analysis (327), or 3) evaluating differentially methylated regions (DMRs) rather than individual CpG sites to combat multiple test correction (328,329).In contrast, the more severe pathology underlying EOPE results in many readily detected DNAm changes.  However, even in the case of EOPE, where many large changes in DNAm are 82  identified and can be validated based on a nominal p-value<0.05 in an independent cohort, those sites selected for having the highest magnitude of change rarely showed the same degree of difference in the second cohort.   In conclusion, whether in the context of PE, or other heterogeneous diseases, DNAm may be a useful tool to independently and qualitatively classify pathological groups prior to analysis.  This method may aid in creating more robust prediction algorithms for predicting pathology versus controls. Further studies with larger sample sizes and additional clinical variables are needed to confirm the presence of multiple subtypes of placental-mediated PE and identify what is driving these different subtypes.     83  Chapter 5:!Placental telomere length decline with gestational age differs by sex and TERT, DNMT1, and DNMT3A DNA methylation  5.1! Introduction Telomeres, repetitive sequences located at the end of each chromosome, protect chromosome ends from degradation and end-to-end fusion. As DNA polymerase cannot synthesize DNA in the 3’- 5’ direction, telomere length (TL) decreases with every cell division in the absence of telomerase activity. For this reason, TL has been viewed as a mitotic clock, whereby shorter telomeres are associated with advanced biological age (330). However, other factors in addition to age/cell-division can influence TL.   Initial TL depends on the average length inherited in the gametes, which is positively affected by paternal age (331). It also depends on telomerase activity, the enzyme responsible for telomere maintenance and lengthening. Telomerase is composed of a subunit with telomerase reverse transcriptase (TERT) activity and a non-coding RNA encoded by telomerase RNA component (TERC) and having sequence complementary to the hexameric telomeric repeat (TTAGGG) (332). Telomerase is active in the gametes, early embryo, and placenta but is mostly silent in somatic tissues after birth (333). Telomerase activity is also positively influenced by estrogen level, such that high estrogen is associated with longer telomeres (334). This relationship may be explained by presence of an estrogen response element (ERE) in the promoter of the telomerase reverse transcriptase gene TERT (334). Guanine repeats within the telomeric region are 84  susceptible to oxidation, causing them to be a genomic location susceptible to DNA damage (148). The resulting DNA repair process can also result in telomere shortening (335).   While few genetic variants have been associated with TL, (336), mutations in TERC can lead to Dykeratosis Congenita, a premature aging disease (337,338,339) and polymorphisms in TERC have been linked to changes in TL (340). As little of the estimated 30-80% TL heritability can be explained by these or other variants, it was proposed that epigenetic variation may additionally influence TL (341). Mutations in DNA methyltransferase (DNMT) genes, responsible for establishing and maintaining DNAm across the genome, have also been associated with aberrant TL (303), providing evidence for a role of DNAm in telomere maintenance. Mutations in DNMT3B lead to immunodeficiency-centromeric instability-facial anomalies syndrome (342), characterized by hypomethylation at subtelomeric regions, which results in short TL (343). Additionally, knock out studies of DNMTs in mice result in abnormal telomere elongation (344).   A few studies have reported reduced TL in placentas associated with adverse outcomes compared to controls (149,150,152). Oxidative stress is thought to be increased in placentas and fetuses associated with preeclampsia (PE) and/or intrauterine growth restriction (IUGR), due to inefficient remodeling of the uterine spiral artery and resulting in reduced blood and oxygen flow to the placenta and fetus (4). This increased oxidative stress may result in reduced TL in placentas from pregnancies complicated by PE and/or IUGR (150). However, very little is known about TL dynamics in the human placenta during gestation.  85  In this study, TL was evaluated in a set of 140 placentas from healthy and complicated pregnancies, using quantitative PCR (qPCR). The effects of factors that may influence placental TL in healthy pregnancies including gestational age, fetal sex, maternal age, and birth weight was first assessed. The relationship of TL with pregnancy complications was evaluated after considering these covariates. DNAm was measured in a subset of 58 placentas using the Illumina Infinium HumanMethylation450 BeadChip array (450K). TL was first compared to DNAm at biologically relevant genes (TERT, TERC and DNMTs) and then to all sites targeted by the array.   5.2! Methods 5.2.1! Sample Information To study TL across gestation, 92 placental samples were used that spanned from 6 weeks gestation to term (Table 5.1). Sources included elective termination (N= 17), preterm birth due to various etiology with minimal placental involvement (e.g. incompetent cervix, spontaneous premature rupture of the membranes, and preterm labor) (N= 55) and healthy term deliveries (N= 20). To compare TL between pathological groups to control placentas, only the subgroup of 59 preterm control placentas from viable pregnancies (25.7-40.3wks GA) was used.   Table 5.1. Placental samples and clinical data for which telomere length was measured. Group N= Gestational Age Range (weeks) Sex (F:M) Preterm (<24wks) 33 6.4-23.9 19:14 Controls  (>24wks) 59 25.7-40.3 27:32 EOPE 21 24.9-37.6 8:13 LOPE 18 34.6-41.3 9:9 IUGR 9 35.4-38.0 5:4   86   5.2.2! Telomere length measures Average relative TL was measured in 92 control, 21 EOPE, 18 LOPE, and 9 IUGR placentas by qPCR as described by Cawthon et al. (2002) (345), with several modifications described in Hanna et al. (2009) (346). Average TL is reported as a Telomere/Single gene (T/S) ratio which is the amplification of the telomere repeat region (T) relative to the amplification of a single copy housekeeping gene (S), 36B4. Samples were run in triplicate on 96-well plates. Each plate contained 23 samples, a standard curve made up of reference DNA, serially diluted from 20ng to 0.625ng, a no-template control, and both a long and short telomere reference control. 7.0ng of DNA was loaded into each well. To make certain that only our targeted amplicons were being amplified, dissociation melting curves were run after each sample. In cases where the standard deviation (SD) between the triplicate sample were >0.2, the sample was removed and re-run on the next plate. Additionally, samples which had a coefficient of the variance (CoV) >0.15 were also removed and re-run. Each sample was replicated (in triplicate) on a separate plate to ensure that T/S values were reliable. The average of these two independent T/S values is reported for each sample. In cases where the CoV between the two T/S values was >0.15, the sample was re-run and the average between the two most similar runs is reported.  5.2.3! DNA methylation analysis DNAm was measured on the Illumina Infinium HumanMethylation450 (450K) array for 20 EOPE, 17 LOPE, 8 IUGR and 13 control placentas for which TL was available. We first examined DNAm alterations at six candidate genes (TERT, TERC, DNMT1, DNMT3a, and 87  DNMT3b) relevant to telomerase and DNAm establishment and maintenance, then performed a genome-wide approach utilizing the entire array.  5.2.4! Statistical analysis Linear modeling was used to assess the change in TL with gestational age, maternal age, birth weight and pathological groups. These analyses used only the preterm placentas >24 weeks gestation (N=59). Two-tailed t-tests were used to determine whether placental TL differed between fetal sexes. Linear modeling, accounting for fetal sex and gestational age, was used to compare mean TL in pathological groups to the control group. A further subset of controls for which 450K array data was available (28-38wks gestation) was used to evaluate the association between DNAm and TL. To estimate global DNAm, we averaged the DNAm measure of all 441,093 probes in our analysis for each of our samples to yield a global average DNAm value. For the array-wide analysis, we set a false discovery rate (FDR) <0.05 and a change in DNAm (Δβ) >0.05 with every 1.0 change in T/S value. All statistical analysis was performed in R, version 3.2.5(347).  5.3! Results 5.3.1! Telomere length is associated with gestational age and fetal sex TL dynamics spanning all three trimesters of pregnancy showed shorter TL in male fetuses compared to female fetuses, independent of gestational age (Figure 5.1, R=-0.43 p=0.001). Maternal age was not associated with TL (Figure 5.2a; R=-0.001, p=0.92), nor was birth weight (Figure 5.2b; R=-0.008, p=0.36). Gestational age at delivery was negatively associated with TL (Figure 5.2c; R=-0.04, p=0.02). On average, female placentas had longer telomeres than male 88  placentas, even when gestational age was considered (Figure 5.2d, p=0.004).  Based on this analysis, both gestational age and fetal sex were adjusted for in subsequent analyses.    !Figure 5.1. Fetal sex is associated with telomere length. Taking into account gestational age, fetal sex is significantly associated with telomere length (R=-0.43; p=0.001). Average telomere length as a T/S ratio is on the y-axis and gestational age (weeks) is on the x-axis.89   Figure 5.2. Variables assessed as potential confounders with telomere length.  In each graph, average telomere length, measured as a T/S ratio, is depicted on the y-axis in comparison to A) maternal age, B) birth weight, measured as standard deviation (SD), C) gestational age, and D) fetal sex.90   5.3.2! Telomere length does not differ between EOPE, LOPE, IUGR and controls Adjusting for gestational age and fetal sex, there were no significant differences in placental TL between EOPE, LOPE, or IUGR compared to controls (Figure 5.3), though TL tended to be shorter in EOPE placentas compared to controls (R=-0.25, p=0.10). TL was significantly longer in LOPE placentas as compared to EOPE placentas (R=0.48, p=0.02). 91   !Figure 5.3. Boxplots depicting telomere length (T/S ratio) between pathological groups.  EOPE, early-onset preeclampsia; LOPE, late-onset preeclampsia; IUGR, intrauterine growth restriction. † p=<0.1, * p<0.05.92   5.3.3! Telomere length is associated with DNA methylation at TERT, DNMT1, and DNMT3A We assessed whether DNAm at TERT or TERC was associated with TL. There were 99 and 15 probes on the 450K array that are within TERT and TERC, respectively. Assessing each CpG site independently, adjusting for gestational age and fetal sex, no association was found between TERC DNAm and average TL. An association between DNAm at a CpG site upstream of the TERT promoter CpG island and a CpG site in the gene body of TERT was significant (cg01934390 Δβ=0.053 for every 1.0 T/S unit, p=0.003, cg11832804 Δβ=0.052 for every 1.0 T/S unit, p=0.01), whereby longer TL was associated with increased DNAm at these sites (Figure 5.4a). Overall, 19 of the 99 TERT probes were nominally significant without an additional Δβ cutoff, of which 18 showed increased DNA methylation associated with longer telomeres  DNAm of two CpG sites within DNMT1 were associated with TL, although the change in DNAm per 1.0 unit change in T/S ratio was slightly below our arbitrary cutoff (cg07267628 Δβ=0.043, p=0.007, cg26538782 Δβ=0.044, p=0.03, Figure 5.4b). Cg07267628, is located in DNMT1, downstream of the transcriptional start site and in the first exon of an alternate transcript of DNMT1, while, cg26538782, is located within an enhancer, upstream of DNMT1. DNAm at a CpG within the gene body of DNMT3A and upstream of the promoter region of an alternate transcript was also associated with TL (cg11779362, Δβ=0.051, p=0.004, Figure 5.4c). There was no association between DNAm within DNMT3B and TL. 93  !Figure'5.4.'Sites'in'TERT,'DNMT1'and'DNMT3A'with'association'between'DNA'methylation'and'telomere'length.'Change!in!DNAm!(∆β)!with!every!1.0!change!in!T/S!ratio!is!on!the!y?axis.!CpG!site!(IlluminaID!name)!in!genomic!order!is!on!the!x?axis.!Sites!marked!with!an!X!meet!false!discovery!rate!(FDR)!<0.05.!Gene!map!is!depicted!below!the!line!graph,!where!green!is!the!CpG!islands,!red!in!the!promoter!region,!yellow!in!enhancer!and!blue!is!CTCF!binding!sites.!Alternative!transcripts!are!in!grey.DNMT3A DNMT1TERT 94  We further investigated if DNAm at the sites that associate with TL (TERT, DNMT1, and DNMT3A), differed between EOPE, LOPE, IUGR and controls. Adjusting for fetal sex and gestational age, DNAm at cg01934390 within TERT differed between EOPE and controls (p=0.02), and EOPE and LOPE (p=0.04, Figure 5.5Ai). No differences in DNAm were observed at cg11832804 within TERT (Figure 5.5Aii).  Within DNMT1, DNAm at cg07267628 differed between EOPE and controls (p=0.001), LOPE and controls (p=0.009), and IUGR and control (p=0.001) (Figure 5.5Biii) groups. DNAm at cg26538782, within DNMT1 only differed between EOPE and controls (p=0.03, Figure 5.5Biv). Similarly, DNMT3A only showed significant changes between EOPE and controls (p=0.03, Figure 5.5Cv).   95  !Figure 5.5. DNA differences within TERT, DNMT1, and DNMT3A. A) DNA methylation between placental pathology groups within TERT at i) cg01934390 and ii) cg11832804. B) DNA methylation between placental pathology groups within DNMT1 at iii) cg07627628 and iv) cg26538782. C) DNA methylation between placental pathology groups within DNMT3A at site cg11779362. 96  5.3.4! Genome-wide DNA methylation analysis Taking an array-wide approach using all the probes on the array that passed quality control (n=441,093), no CpG sites were associated with TL using a threshold of a FDR <0.05 and Δβ >0.05 for every 1.0 T/S unit. The Δβ cutoff was used to enrich for sites with larger changes in DNAm which may have more biological relevance but does not mean sites not meeting the cutoff are any less meaningful.  Despite changes in DNAm within DNMTs, we observed no difference in average global DNAm (average DNAm across all probes on the 450K array) with varying TL (R=-0.003, p=0.79).  5.4! Discussion TL has been reported to differ between the sexes in a number of tissues (348,349); however, few studies assessed the relationship between fetal sex and TL in the placenta or assessed variation in placental TL between fetal sexes over gestation (152,153). We observed longer telomeres in female placentas compared to male placentas, which may be related to differences in the levels of estrogen, a hormone known to activate telomerase (334). Gielen et al. (2014) (153) previously found no difference in placental TL between male and female fetuses within dizygotic twin pairs. However, Benetos et al. (2014) observed that leukocyte TL in female fetuses that were part of a male-female twin pair had shorter TL compared to those from a female-female twin pair, and similar TL to their male twin. They proposed that the hormone environment in utero influences TL independent of biological sex (349).   After accounting for fetal sex and gestational age, we did not confirm previous reports (152) (350) of shorter TL in placentas from pregnancies complicated by EOPE, LOPE or IUGR. These 97  later studies did not correct for sex differences, which may have affected the results as preeclampsia and IUGR are generally more common in male than female pregnancies. Gielen et al. (2014) (153) also failed to observe a difference in PE and IUGR associated placentas compared to controls. We did however, observe a trend to shorter telomeres in EOPE, which was significant in comparison to LOPE (p=0.01). Our power to detect a difference may have been limited by small sample size.  Telomere attrition may be rescued by telomerase activity. As TERT expression is the rate-limiting factor to telomerase activity (351), it is interesting that we found an association between shorter TL and less DNAm within the TERT gene body and upstream of the transcriptional start site. Moreover, without using a biological cutoff, 19 of the 99 probes within TERT showed nominally significant alterations in DNAm associated with TL. In contrast to many genes, hypermethylation within at least part of the TERT promoter is associated with TERT expression, thereby leading to increased telomerase activity (352). However, epigenetic regulation of the TERT promoter is complex; DNAm at a CTCF binding site near the promoter region is necessary for TERT transcription (to prevent transcription factors from binding to this site), while selective hypomethylation of part of the promoter is required to allow the transcription factors that aid in TERT transcription to bind(352).  Changes in DNAm observed throughout the entire TERT gene region will need further investigation to elucidate the biological relevance of these changes.  DNAm alterations in DNMTs (DNMT1 and DNMT3a) further support the previous suggestions that epigenetic variation may contribute to TL (344,353). DNAm changes were observed at cg11779362 upstream of the promoter region to an alternative transcript of DNMT3a. This 98  alterative transcript, referred to as DNMT3a2 is highly expressed in embryonic stem cells and is thought to preferentially methylate euchromatin domains compared to DNMT3a which preferentially methylates heterochromatin domains (354). However, it is unclear whether these alterations in DNAm have biological consequences and would subsequently affect telomere biology. Despite the DNAm changes in DNMT1 and DNMT3A, we failed to see alterations in average array-wide DNAm; though this may be related to specific gene expression patterns of these genes in the trophoblast, not genome-wide effects (355).   Understanding placental telomere biology and how it relates to normal and abnormal placental development remains of continued interest. Although this study indicates no major difference in TL with EOPE, LOPE, and IUGR, TL has been suggested to act as a biological clock, potentially signaling the onset parturition (356) and may have implications for pregnancy complications.  99  Chapter 6:!Discussion 6.1! Summary of Dissertation In this thesis I investigated genetic and epigenetic profiles in placentas from pregnancies affected by placental insufficiency conditions to assess biomarker utility of these molecular marks. These analyses included i) characterizing DNA methylation (DNAm) profiles, ii) assessing whether DNAm in the term placenta can reflect protein levels in maternal blood earlier in gestation thus, whether DNAm could be used to identify novel protein biomarkers, and iii) measuring placental telomere length in preeclampsia (PE) and/or intrauterine growth restriction (IUGR) placentas. In this discussion, I will review the main findings of this dissertation and their significance, outline the strengths and limitations of these studies, discuss reproducibility and responsible reporting of differential methylation, and future directions for the field of placental insufficiency.  !6.2! Significance and contribution In chapter 3, I assessed whether DNAm in the term placenta reflects protein levels in maternal blood during the first, second and third trimester. While DNAm in the term placenta was correlated to Inhibin A (INHA) protein levels in maternal blood in second trimester and Fibronectin (FN1) protein levels in the third trimester, it did not correlate in the first trimester pregnancy associated plasma protein A (PAPPA) levels. This study suggests a link between placental DNAm at delivery to earlier changes in gene expression and protein levels for some markers in maternal blood; supporting the hypothesis that DNAm may represent a ‘cellular memory’ of altered molecular changes occurring at earlier developmental time points (357). While it is possible to utilize DNAm to identify novel protein biomarkers in maternal blood earlier in gestation, this may not be possible for first trimester biomarkers as DNAm in the term 100  placenta may not reflect gene expression or protein expression alterations in the first trimester. A number of confounding factors can also inhibit the utility of DNAm to reflect protein levels in maternal blood (Figure 3.4). This study provides evidence that DNAm at delivery may correlate with altered protein levels earlier in gestation, providing an alternate route of investigation in identifying biomarkers with the potential of predicting pregnancies at risk of developing placental insufficiency conditions.   Chapter 4 characterizes the placental DNAm profiles of early-onset PE (EOPE), late-onset PE (LOPE), and IUGR in comparison to preterm birth and term birth controls. I observed widespread DNAm alterations in EOPE compared to preterm controls, but not in LOPE or IUGR compared to term controls. There were 599 CpG sites identified that validated in an independent cohort. Clustering on these validated sites revealed 3 distinct clusters in both cohorts. Interestingly, the LOPE cases with other indications of placental insufficiency (e.g. fetal growth restriction) clustered with the EOPE cases, which are more commonly associated with placental insufficiency. Our data suggested that DNAm profiling may be valuable as an independent tool to help classify the hard to diagnose cases being diagnosed with PE in the grey zone between calling a case EOPE or LOPE. This was the first study to assess genome-wide DNAm and validate differentially methylated sites in an independent cohort, in EOPE, LOPE and IUGR placentas.   Chapter 5 assessed telomere length dynamics in the human placenta across gestation and between placental insufficiency conditions and controls. It was found that telomere length decreased over gestational age, and that telomere length was shorter in male placentas compared 101  to female placentas. While telomere length has been reported to be shorter in PE and IUGR placentas (150,152,358), I was!unable to confirm this result when correcting for both gestational age and fetal sex; although telomere length did trend toward shorter length in EOPE placentas. This was the first study to assess telomere length dynamics in the placenta across all three trimesters of pregnancy.   In this dissertation I demonstrate the utility of DNAm studies for identifying novel biomarkers or epigenetic signatures capable of predicting pregnancies at risk of placental insufficiency conditions. While there are many obstacles that will need to be overcome, such as how to identify placental specific DNAm in maternal blood, DNAm studies hold much promise for assessing placental health. Placental DNAm studies can provide insight into normal placental development and pathogenic mechanisms associated with placental insufficiency conditions. Additionally, placental DNAm studies can also shed light onto the ramifications of environmental exposures to placental and fetal health (359). These DNAm alterations can be taken to identify and refine prediction algorithms for studying disease and exposure phenotypes.   !6.3! Strengths and limitations Evaluating genetic and epigenetic placental profiles in these studies is underpowered to detect small differences. I have tried to mediate this limitation by setting a Δβ threshold >0.10 (10% points difference between cases and controls) to enrich for sites more likely to be biologically relevant, and not within the normal variability of a site. Additionally, while clinical information on gestational age, fetal sex, birth weight, maternal age, etc. were collected to try to mediate any confounding relationships between telomere length or DNAm and placental insufficiency, it is 102  possible that there are other variables influencing both telomere length and DNAm that have not been reported or assessed. Many of the limitations of this dissertation are limitations of DNAm studies in general and will be further discussion in section 6.6. These studies are limited to investigating DNAm at sites that are present on the Illumina Infinium HumanMethylation450 (450K) array. The 450K probes are enriched for promoters and high-density CpG regions of the genome, which have been more often associated with altered gene expression. However, this limits any assessment of altered DNAm at many intergenic and low CpG density regions in association with PE and IUGR. Additionally, the annotation of enhancer and other regulatory regions were determined by ENCODE and Illumina on non-placental cell types (360); without further studying of basic placental biology, it will be difficult to fully elucidate the consequences of DNAm alterations at these regions in the placenta. The 450K methylation array is enriched for specific genomic regions, which in some sense may limit these studies. By restricting CpG sites to be investigated to sites more likely to have a biological effect makes data interpretation somewhat easier and reduces the number of multiple test corrections needed, increasing power of the study. The microarray is a highly validated technique that correlates well to other methods of measuring DNAm (273), and correlates well between batches of microarrays (271). As it is such a widely used method and open source data is available publically, using 450K or similar platforms make it easier to compare to results from other studies. A limitation with this data is that probes present on the 450K array are enriched for developmental pathways, transcription factor binding and, cell differentiation gene ontology terms (361). This adds another layer of complexity when interpreting gene ontology analysis results from 450K array results. To mediate this effect, gene ontology analysis using ermineJ software was completed with a 450K background annotation. 103   The placenta is an extremely heterogeneous tissue, comprised of many different cell types. Variation in DNAm across the placenta has been reported (238). To account for this variation in our studies, multiple sites in the placenta were sampled and DNA was pooled together in equal amounts to give a more accurate representation of the placenta as a whole. Details on this can be found in chapter 2.   6.4! Approaches and considerations in reproducibility of DNA methylation studies Many studies have reported altered DNAm at CpG sites within the placenta from pregnancies complicated by PE (13,16,246,322) and IUGR (50,53,323). Although findings have been reported in numerous studies, only one study validated their findings, using the same technology, in an independent cohort (240). As DNAm is susceptible to technical and batch effects, validating hits is imperative to understanding what is truly associated with pathology. In chapter 4, 35.1% (N=599) sites were found to be differentially methylated between EOPE and preterm controls in both the discovery and validation cohorts using validation criteria of a nominal p-value<0.05 and change in DNAm in the same direction as the discovery cohort. The extent of validation, however, is dependent on the initial criteria chosen to define ‘hits’, the criteria for validation, the similarity of the populations of samples, the size of study populations (power to detect changes) and the similarity of processing of samples.  The reproducibility rate depends on how validation is defined. Often top hits in the discovery cohort are compared to the top hits in the validation cohorts using both a statistical and biological (Δβ) threshold. This can be a very stringent approach and may eliminate sites that are concordant 104  between cohorts, but just do not reach the same level of significance in both.  For this study I opted to be less stringent and used a nominal p-value to call a hit validated. However, when I used a Bonferroni-corrected p-value <0.05, validated hits decreased from 599 (35.1%) to 42 (2.5%). Although reducing the data to 42 hits may have some benefits, it can become problematic when it reduces the data so much that it is biologically meaningless.  False discovery rate (FDR) is a statistical measure of significance often used to correct for multiple testing. I used FDR in the discovery cohort as it corrects for multiple test corrections but is less stringent than the Bonferroni test correction. However, it should be pointed out that an FDR<0.05 is not the same as a p-value<0.05. Where a p-value<0.05 is interpreted as 5% of the total tests would be a false positive, an FDR is calculated based on the p-value distribution of a specific dataset and an FDR<0.05 is interpreted as 5% of the identified positive hits are actually false positive.   Determining what statistical and biological thresholds to use will depend on the question of the study. Less stringent criteria might be used if the goal is simply to characterize a pathology, while more stringent criteria may be important if the purpose is to develop a panel of biomarkers for screening. An alternative approach may be to reduce the dimensions in the data by 1) removing probes that are non-variable removing across all cell type (326), 2) or focusing on alterations in pathway modules, as in weighted gene co-expression network analysis(327) or 3) evaluate differentially methylated regions (DMRs), rather than individual CpG sites as is often used to combat multiple test correction (328,329).  When studying complex multifactorial diseases, reproducibility is likely to be increased by rigorously defining and classifying the disease of interest to create more homogenous groups. In 105  many cases, as with PE and IUGR, disease phenotypes consist of a variety of subtypes. Studies tend to focus on identifying homogenous subtypes of pathologies to increase the likelihood that molecular changes will be identified. However, the selection of control samples should also be considered. In this dissertation, to match gestational age of controls with EOPE, I needed to use cases of preterm birth, which can include a variety of abnormal etiologies. In the discovery cohort preterm births were selected to represent a range of causes in order to reduce the chance that changes observed were due to pathology associated with preterm birth rather than EOPE. Some studies will opt to not use preterm births as controls, comparing only to healthy term placentas (13,53,244). However, this introduces the limitation that the pathology and control groups are completely confounded by gestational age. A potential solution would be to identify sites that are differentially methylated between EOPE and preterm controls and also differentially methylated between EOPE and term controls. While this is likely going to eliminate some true differentially methylated sites, we can be more confident in the smaller subset of sites that is identified with this method. Unfortunately in our dataset this was not an option as preterm and term controls were run in separate batches and therefore batch effects might still have been an issue.    6.5! Understanding limitations and responsible reporting of results in EWAS studies  While epigenetic-wide association studies (EWAS) have become very common for assessing changes in response to perturbations, there are still many limitations that need to be considered when interpreting the results. I will be discussing limitations specifically in the context of DNAm, although many of these limitations are also relevant to other epigenetic marks.  The relationship between DNAm and gene expression has been widely studied and is thus more 106  intuitive than other epigenetic marks, however, the interpretability and biological meaning of a change in DNAm is relatively unclear. DNAm changes are often interpreted as causation for a perturbation or pathogenesis (DNAm alterations causing phenotype); however, DNAm can change in response to or as an adaptation and therefore secondary to the perturbation (Phenotype causes DNAm change). Moreover, as DNAm patterns differ widely between cell types (210), DNAm alterations may reflect altered cell composition in the tissue of interest. Variance due to altered cell type proportions is often removed from DNAm studies (362,363,364,365), as it is thought to be a confounding variable of phenotype. However, more thought should be taken before removing cell type proportion variance out of the data, as cell type proportion may actually be causal to the phenotype of interest. In a recent review by Lappalainen and Greally (2017), the term polycreodism is used to define cell type proportion variation that arises in response to an exposure or perturbation during development (357). Therefore, DNAm changes that reflect altered cell type proportions may actually be biologically relevant and related to the phenotype of interest. In the context of EOPE, most studies focus on DNAm altering gene expression pathways associated with PE (e.g. related to known effects such as oxidative stress and altered angiogenesis (26,67)). However, DNAm changes may reflect altered cell type proportions related to PE pathology, such as decreased proliferation of extravillous trophoblast cells or alterations to the rate of trophoblast proliferation (324).  As cell-type specific profiles have not been developed for all placental cell types, it is not possible to use the DNAm profile to estimate cell proportions, as has been applied to blood (213). While reference-free methods for deconvolution of cell proportions have been developed (325), these methods remove variance within the data attributed to cell composition but cannot inform us of what cell types specifically are altered in EOPE. Even when removing variability due to cell type with current methods, cell 107  type variability will likely persist, as it is not clear exactly how many cell types are present within a specific tissue and whether cell type is a discrete categorical variable or a continuous variable defined by alterations in molecular profiles (357). Another limitation of EWAS studies is often there is no hypothesis driving the study, we are searching for differences in DNAm with no clear hypothesis as to whether these changes are causal or secondary to the phenotype of interest. In the case of studies attempting to identify biomarkers for a phenotype, a clear hypothesis is less important as biological meaning and interpretation is not a priority over the DNAm mark’s predictive capabilities. This is discussed in further detail in Lappalainen and Greally (2017) (357).!  Many variables, independent of the phenotype of interest, influence DNAm, and need to be accounted for in the experimental design. While variables such as age and sex are almost always adjusted for in DNAm studies, it is often forgotten how intertwined epigenetics is with DNA sequence. Methylation quantitative trait loci (mQTL) are CpG sites where the DNAm status is influenced by nearby single nucleotide polymorphisms (SNPs)(366). As such, an individual’s genetic ancestry can impact DNAm status across the genome (179,367). A challenge in epigenetic studies is that reported ethnicity does not always accurately represent an individual’s genetic ethnicity (368). To account for this, numerous studies have utilized ancestry informative markers (AIMs) to more accurately correct for ethnicity within their studies (369). DNAm markers have also been developed to predict ethnicity in studies where ethnicity may not have been available (370).  However, genetic variation associated with ancestry is unlikely to explain all genetic variability present in epigenetic studies. This likely results in many DNAm alterations reported that are actually genetic effects on phenotype rather than alterations due to 108  environmental exposure or disease phenotype.  For example, genetic variants associated with other factors such as hypertension, body mass index (BMI), or diabetes, may affect DNAm. Since these are all also risk factors for PE, we could identify changes in DNAm that actually reflect genetic variation associated with the risk factor rather than PE itself.  6.6! Future directions of preeclampsia and intrauterine growth restriction prediction Molecular profiling (transcriptomics, DNA methylation, miRNA profiling, proteomics) are all paving the way for understanding the pathogenesis of PE and IUGR, and the multiple subtypes that make up the broader disease. Further research in these areas will be needed to fully understand placental insufficiency conditions. Integrating these different data types into multiple ‘omics’ studies will create a clearer representation of the relationships between pathway systems and the development or PE and IUGR. In particular, determining if altered DNAm in placentas associated with PE is a cause or a consequence to the pathology of PE, and whether the DNAm changes are actually reflecting polycreodism, altered cell type proportions in the placenta, will be a major area of research in the coming years.   While DNAm alterations are reported in PE placentas (Supplementary Table A.1), these DNAm alterations can only be used as biomarkers i) through placental sampling, which would be an invasive test not applicable to the general population, or ii) if these DNAm are detectable in the maternal serum. As outlined in chapter 3, there are many obstacles that may prevent these DNAm alterations from being detectable in maternal blood during gestation. However, studies utilizing the global hypomethylation and promoter hypermethylation profile of cancer have used methylated DNA immunoprecipitation (MeDIP) analysis to capture cell free tumour DNA in the 109  patient’s serum (371,372). The same technique has been utilized to capture cell free placental DNA in maternal serum for identification of chromosomal abnormalities (373,374). Future PE and IUGR DNAm profiling for the intent of biomarker identification should consider utilizing this technique, or similar method to characterize the DNAm profile in maternal blood. This will eliminate the limitations of placental DNAm profiling, as DNAm changes will be detectable in maternal blood, and the test is non-invasive, making it available to all pregnancies, and easily incorporated into the maternal serum screen.  110  Bibliography 1. Hutcheon J.A., Lisonkova S. and Joseph K. (2011) Epidemiology of pre-eclampsia and the other hypertensive disorders of pregnancy. Best practice & research Clinical obstetrics & gynaecology, 25, 391-403. 2. Jebbink J., Wolters A., Fernando F., Afink G., van der Post J. and Ris-Stalpers C. (2012) Molecular genetics of preeclampsia and HELLP syndrome—a review. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease, 1822, 1960-1969. 3. Magee L.A., Helewa M., Moutquin J.M., von Dadelszen P., Hypertension Guideline Committee and Strategic Training Initiative in Research in the Reproductive Health Sciences (STIRRHS) Scholars (2008) Diagnosis, evaluation, and management of the hypertensive disorders of pregnancy. J. Obstet. Gynaecol. Can., 30, S1-48. 4. Steegers E.A., von Dadelszen P., Duvekot J.J. and Pijnenborg R. (2010) Pre-eclampsia. The Lancet, 376, 631-644. 5. American College of Obstetricians and Gynecologists (2014) Hypertension in pregnancy.Washington, DC: American College of Obstetricians and Gynecologists; 2013, . 6. Lowe S.A., Brown M.A., Dekker G.A., Gatt S., McLINTOCK C.K., McMAHON L.P., Mangos G., Moore M.P., Muller P. and Paech M. (2009) Guidelines for the management of hypertensive disorders of pregnancy 2008. Australian and New Zealand Journal of Obstetrics and Gynaecology, 49, 242-246. 111  7. Tuffnell D., Shennan A., Waugh J. and Walker J. (2006) The management of severe pre-eclampsia/eclampsia. London (UK): Royal College of Obstetricians and Gynaecologists, 11. 8. National Collaborating Centre for Women's and Children's Health (UK) (2008) . 9. Tranquilli A., Dekker G., Magee L., Roberts J., Sibai B., Steyn W., Zeeman G. and Brown M. (2014) The classification, diagnosis and management of the hypertensive disorders of pregnancy: A revised statement from the ISSHP. Pregnancy Hypertension: An International Journal of Women's Cardiovascular Health, 4, 97-104. 10. Fokom!Domgue J. and Noubiap J.J.N. (2015) Diagnosis of hypertensive disorders of pregnancy in Sub!Saharan africa: A poorly assessed but increasingly important issue. The Journal of Clinical Hypertension, 17, 70-73. 11. Ingaramo R., Alfie J., Bellido C., Bendersky M., Carbajal H. and Colonna L. Guías de la Sociedad Argentina de Hipertensión Arterial para el diagnóstico, estudio, tratamiento y seguimiento de la hipertensión arterial.Buenos Aires: Sociedad Argentina de Hipertensión Arterial; 2011 [citado 11 Jun 2015], . 12. Stillbirth Collaborative Research Network Writing Group (2011) Causes of death among stillbirths. JAMA, 306, 2459-2468. 13. Anton L., Brown A.G., Bartolomei M.S. and Elovitz M.A. (2014) Differential methylation of genes associated with cell adhesion in preeclamptic placentas. PLoS One, 9, e100148. 112  14. Yuen R.K., Peñaherrera M.S., von Dadelszen P., McFadden D.E. and Robinson W.P. (2010) DNA methylation profiling of human placentas reveals promoter hypomethylation of multiple genes in early-onset preeclampsia. European Journal of Human Genetics, 18, 1006-1012. 15. Hogg K., Blair J.D., von Dadelszen P. and Robinson W.P. (2013) Hypomethylation of the LEP gene in placenta and elevated maternal leptin concentration in early onset pre-eclampsia. Mol. Cell. Endocrinol., 367, 64-73. 16. Sundrani D.P., Reddy U.S., Joshi A.A., Mehendale S.S., Chavan-Gautam P.M., Hardikar A.A., Chandak G.R. and Joshi S.R. (2013) Differential placental methylation and expression of VEGF, FLT-1 and KDR genes in human term and preterm preeclampsia. Clinical epigenetics, 5, 6. 17. Nelson D.B., Ziadie M.S., McIntire D.D., Rogers B.B. and Leveno K.J. (2014) Placental pathology suggesting that preeclampsia is more than one disease. Obstet. Gynecol., 210, 66. e1-66. e7. 18. Stark M.W., Clark L. and Craver R.D. (2014) Histologic differences in placentas of preeclamptic/eclamptic gestations by birthweight, placental weight, and time of onset. Pediatric and Developmental Pathology, 17, 181-189. 19. Ogge G., Chaiworapongsa T., Romero R., Hussein Y., Kusanovic J.P., Yeo L., Kim C.J. and Hassan S.S. (2011) Placental lesions associated with maternal underperfusion are more frequent in early-onset than in late-onset preeclampsia. J. Perinat. Med., 39, 641-652. 113  20. Leavey K., Bainbridge S.A. and Cox B.J. (2015) Large scale aggregate microarray analysis reveals three distinct molecular subclasses of human preeclampsia. PLoS One, 10, e0116508. 21. Leavey K., Benton S.J., Grynspan D., Kingdom J.C., Bainbridge S.A. and Cox B.J. (2016) Unsupervised placental gene expression profiling identifies clinically relevant subclasses of human preeclampsia. Hypertension, 68, 137-147. 22. Blair J.D., Yuen R.K., Lim B.K., McFadden D.E., von Dadelszen P. and Robinson W.P. (2013) Widespread DNA hypomethylation at gene enhancer regions in placentas associated with early-onset preeclampsia. Mol. Hum. Reprod.,19,697-708. 23. Yuen R.K., Peñaherrera M.S., von Dadelszen P., McFadden D.E. and Robinson W.P. (2010) DNA methylation profiling of human placentas reveals promoter hypomethylation of multiple genes in early-onset preeclampsia. European Journal of Human Genetics, 18, 1006-1012. 24. Hui D., Okun N., Murphy K., Kingdom J., Uleryk E. and Shah P.S. (2012) Combinations of maternal serum markers to predict preeclampsia, small for gestational age, and stillbirth: A systematic review. J. Obstet. Gynaecol. Can., 34, 142-153. 25. Dugoff L. and Society for Maternal-Fetal Medicine (2010) First- and second-trimester maternal serum markers for aneuploidy and adverse obstetric outcomes. Obstet. Gynecol., 115, 1052-1061. 26. Powers R.W., Roberts J.M., Plymire D.A., Pucci D., Datwyler S.A., Laird D.M., Sogin D.C., Jeyabalan A., Hubel C.A. and Gandley R.E. (2012) Low placental growth factor across 114  pregnancy identifies a subset of women with preterm preeclampsia: Type 1 versus type 2 preeclampsia? Hypertension, 60, 239-246. 27. Smets E.M., Visser A., Go A.T., van Vugt J.M. and Oudejans C. (2006) Novel biomarkers in preeclampsia. Clinica chimica acta, 364, 22-32. 28. Tranquilli A.L., Brown M.A., Zeeman G.G., Dekker G. and Sibai B.M. (2013) The definition of severe and early-onset preeclampsia. statements from the international society for the study of hypertension in pregnancy (ISSHP). Pregnancy Hypertension: An International Journal of Women's Cardiovascular Health, 3, 44-47. 29. Lisonkova S. and Joseph K. (2013) Incidence of preeclampsia: Risk factors and outcomes associated with early-versus late-onset disease. Obstet. Gynecol., 209, 544. e1-544. e12. 30. Sibai B., Dekker G. and Kupferminc M. (2005) Pre-eclampsia. The Lancet, 365, 785-799. 31. Sibai B., Ewell M., Levine R., Klebanoff M., Esterlitz J., Catalano P., Goldenberg R. and Joffe G. (1997) Risk factors associated with preeclampsia in healthy nulliparous women. Am. J. Obstet. Gynecol., 177, 1003-1010. 32. Bartsch E., Medcalf K.E., Park A.L., Ray J.G. and High Risk of Pre-eclampsia Identification Group (2016) Clinical risk factors for pre-eclampsia determined in early pregnancy: Systematic review and meta-analysis of large cohort studies. BMJ, 353, i1753. 33. Dekker G. and Sibai B. (2001) Primary, secondary, and tertiary prevention of pre-eclampsia. The Lancet, 357, 209-215. 115  34. Hayashi R.H. (1983) Management of the pregnant patient with chronic hypertension. Clin. Ther., 5, 568-578. 35. Baumwell S. and Karumanchi S.A. (2007) Pre-eclampsia: Clinical manifestations and molecular mechanisms. Nephron Clin. Pract., 106, c72-81. 36. Klatsky P.C., Delaney S.S., Caughey A.B., Tran N.D., Schattman G.L. and Rosenwaks Z. (2010) The role of embryonic origin in preeclampsia: A comparison of autologous in vitro fertilization and ovum donor pregnancies. Obstet. Gynecol., 116, 1387-1392. 37. Einarsson J.I., Sangi-Haghpeykar H. and Gardner M.O. (2003) Sperm exposure and development of preeclampsia. Obstet. Gynecol., 188, 1241-1243. 38. Verwoerd G., Hall D., Grové D., Maritz J. and Odendaal H. (2002) Primipaternity and duration of exposure to sperm antigens as risk factors for pre!eclampsia. International Journal of Gynecology & Obstetrics, 78, 121-126. 39. Ness R.B., Markovic N., Harger G. and Day R. (2004) Barrier methods, length of preconception intercourse, and preeclampsia. Hypertension in pregnancy, 23, 227-235. 40. Rijhsinghani A., Yankowitz J., Strauss R.A., Kuller J.A., Patil S. and Williamson R.A. (1997) Risk of preeclampsia in second-trimester triploid pregnancies. Obstetrics & Gynecology, 90, 884-888. 41. Robinson W., Barrett I., Bernard L., Telenius A., Bernasconi F., Wilson R., Best R., Howard-Peebles P., Langlois S. and Kalousek D. (1997) Meiotic origin of trisomy in confined placental 116  mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. Am. J. Hum. Genet., 60, 917. 42. Robinson W.P., Peñaherrera M.S., Jiang R., Avila L., Sloan J., McFadden D.E., Langlois S. and von Dadelszen P. (2010) Assessing the role of placental trisomy in preeclampsia and intrauterine growth restriction. Prenat. Diagn., 30, 1-8. 43. Magee L.A., von Dadelszen P., Rey E., Ross S., Asztalos E., Murphy K.E., Menzies J., Sanchez J., Singer J. and Gafni A. (2015) Less-tight versus tight control of hypertension in pregnancy. N. Engl. J. Med., 372, 407-417. 44. Roberge S., Villa P., Nicolaides K., Giguère Y., Vainio M., Bakthi A., Ebrashy A. and Bujold E. (2012) Early administration of low-dose aspirin for the prevention of preterm and term preeclampsia: A systematic review and meta-analysis. Fetal. Diagn. Ther., 31, 141-146. 45. Bujold E., Tapp S., Audibert F., Ferreira E., Forest J., Rey E., Fraser W., Chaillet N. and Giguere Y. (2011) Prevention of adverse pregnancy outcomes with low-dose ASA in early pregnancy: New perspectives for future randomized trials. J Obstet Gynaecol Can, 33, 480-483. 46. Rolnik D.L., Wright D., Poon L.C., Syngelaki A., O’Gorman N., de Paco Matallana C., Akolekar R., Cicero S., Janga D. and Singh M. (2017) ASPRE trial: Performance of screening for preterm pre!eclampsia. Ultrasound in Obstetrics & Gynecology, . 117  47. Askie L.M., Duley L., Henderson-Smart D.J. and Stewart L.A. (2007) Antiplatelet agents for prevention of pre-eclampsia: A meta-analysis of individual patient data. The Lancet, 369, 1791-1798. 48. Rolnik D.L., Wright D., Poon L.C., O’gorman N., Syngelaki A., de Paco Matallana C., Akolekar R., Cicero S., Janga D. and Singh M. (2017) Aspirin versus placebo in pregnancies at high risk for preterm preeclampsia. N. Engl. J. Med., 377, 613-622. 49. World Health Organization (2010) World Health Statistics 2010. World Health Organization. 50. Banister C.E., Koestler D.C., Maccani M.A., Padbury J.F., Houseman E.A. and Marsit C.J. (2011) Infant growth restriction is associated with distinct patterns of DNA methylation in human placentas. Epigenetics, 6, 920-927. 51. Guo L., Choufani S., Ferreira J., Smith A., Chitayat D., Shuman C., Uxa R., Keating S., Kingdom J. and Weksberg R. (2008) Altered gene expression and methylation of the human chromosome 11 imprinted region in small for gestational age (SGA) placentae. Dev. Biol., 320, 79-91. 52. Ferreira J.C., Choufani S., Grafodatskaya D., Butcher D.T., Zhao C., Chitayat D., Shuman C., Kingdom J., Keating S. and Weksberg R. (2011) WNT2 promoter methylation in human placenta is associated with low birthweight percentile in the neonate. Epigenetics, 6, 440-449. 53. Roifman M., Choufani S., Turinsky A.L., Drewlo S., Keating S., Brudno M. and Weksberg R. (2016) Genome-wide placental DNA methylation analysis of severely growth-discordant 118  monochorionic twins reveals novel epigenetic targets for intrauterine growth restriction. Clinical Epigenetics, 8, 70. 54. He Z., Lu H., Luo H., Gao F., Wang T., Gao Y., Fang Q. and Wang J. (2016) The promoter methylomes of monochorionic twin placentas reveal intrauterine growth restriction-specific variations in the methylation patterns. Sci. Rep., 6, 20181. 55. Wilson S.L., Blair J.D., Hogg K., Langlois S., von Dadelszen P. and Robinson W.P. (2015) Placental DNA methylation at term reflects maternal serum levels of INHA and FN1, but not PAPPA, early in pregnancy. BMC medical genetics, 16, 111. 56. Koukoura O., Sifakis S., Zaravinos A., Apostolidou S., Jones A., Hajiioannou J., Widschwendter M. and Spandidos D. (2011) Hypomethylation along with increased H19 expression in placentas from pregnancies complicated with fetal growth restriction. Placenta, 32, 51-57. 57. Lambertini L., Lee T.L., Chan W.Y., Lee M.J., Diplas A., Wetmur J. and Chen J. (2011) Differential methylation of imprinted genes in growth-restricted placentas. Reprod. Sci., 18, 1111-1117. 58. Mamelle N., Cochet V. and Claris O. (2001) Definition of fetal growth restriction according to constitutional growth potential. Biol. Neonate, 80, 277-285. 59. De Jong C., Francis A., Van Geijn H. and Gardosi J. (2000) Customized fetal weight limits for antenatal detection of fetal growth restriction. Ultrasound in obstetrics & gynecology, 15, 36-40. 119  60. Maulik D. (2006) Fetal growth compromise: Definitions, standards, and classification. Clin. Obstet. Gynecol., 49, 214-218. 61. Figueras F. and Gardosi J. (2011) Intrauterine growth restriction: New concepts in antenatal surveillance, diagnosis, and management. Obstet. Gynecol., 204, 288-300. 62. Gruslin A. and Lemyre B. (2011) Pre-eclampsia: Fetal assessment and neonatal outcomes. Best practice & research Clinical obstetrics & gynaecology, 25, 491-507. 63. Suhag A. and Berghella V. (2013) Intrauterine growth restriction (IUGR): Etiology and diagnosis. Current Obstetrics and Gynecology Reports, 2, 102-111. 64. Dashe J.S., McIntire D.D., Lucas M.J. and Leveno K.J. (2000) Effects of symmetric and asymmetric fetal growth on pregnancy outcomes. Obstetrics & Gynecology, 96, 321-327. 65. Cox P. and Marton T. (2009) Pathological assessment of intrauterine growth restriction. Best practice & research Clinical obstetrics & gynaecology, 23, 751-764. 66. Rosenberg A. (2008) The IUGR Newborn. Elsevier. 67. Roberts J.M. and Bell M.J. (2013) If we know so much about preeclampsia, why haven’t we cured the disease? J. Reprod. Immunol., 99, 1-9. 68. Roberts J.M. and Hubel C.A. (2009) The two stage model of preeclampsia: Variations on the theme. Placenta, 30, 32-37. 120  69. Redman C., Sargent I. and Staff A. (2014) IFPA senior award lecture: Making sense of pre-eclampsia–two placental causes of preeclampsia? Placenta, 35, S20-S25. 70. Roberts J. and Cooper D.W. (2001) Pathogenesis and genetics of pre-eclampsia. The Lancet, 357, 53-56. 71. Hartley J.D., Ferguson B.J. and Moffett A. (2015) The role of shed placental DNA in the systemic inflammatory syndrome of preeclampsia. Obstet. Gynecol., 213, 268-277. 72. Gude N.M., Roberts C.T., Kalionis B. and King R.G. (2004) Growth and function of the normal human placenta. Thromb. Res., 114, 397-407. 73. Benirschke K., Burton G.J. and Baergen R.N. (2012) Early Development of the Human Placenta. In Pathology of the Human Placenta. Springer, pp. 41-53. 74. Benirschke K., Burton G.J. and Baergen R.N. (2012) Early Development of the Human Placenta. In Pathology of the Human Placenta. Springer, pp. 41-53. 75. Cross J.C., Werb Z. and Fisher S.J. (1994) Implantation and the placenta: Key pieces of the development puzzle. Science, 266, 1508. 76. Staun-Ram E. and Shalev E. (2005) Human trophoblast function during the implantation process. Reproductive Biology and Endocrinology, 3, 56. 77. Bischof P. (2001) Endocrine, paracrine and autocrine regulation of trophoblastic metalloproteinases. Early Pregnancy, 5, 30-31. 121  78. Rurak D.W. (2001) Development and function of the placenta. Fetal growth and development, , 17-43. 79. Sekizawa A., Purwosunu Y., Yoshimura S., Nakamura M., Shimizu H., Okai T., Rizzo N. and Farina A. (2009) PP13 mRNA expression in trophoblasts from preeclamptic placentas. Reproductive Sciences, 16, 408-413. 80. Hemberger M., Dean W. and Reik W. (2009) Epigenetic dynamics of stem cells and cell lineage commitment: Digging waddington's canal. Nature reviews.Molecular cell biology, 10, 526. 81. Rossant J. and Cross J.C. (2001) Placental development: Lessons from mouse mutants. Nature reviews.Genetics, 2, 538. 82. Aplin J.D. (2000) Hypoxia and human placental development. J. Clin. Invest., 105, 559-560. 83. Genbacev O., Zhou Y., Ludlow J.W. and Fisher S.J. (1997) Regulation of human placental development by oxygen tension. Science, 277, 1669-1672. 84. Reynolds L.P. and Redmer D.A. (2001) Angiogenesis in the placenta 1. Biol. Reprod., 64, 1033-1040. 85. Syme M.R., Paxton J.W. and Keelan J.A. (2004) Drug transfer and metabolism by the human placenta. Clin. Pharmacokinet., 43, 487-514. 86. Lager S. and Powell T.L. (2012) Regulation of nutrient transport across the placenta. J. Pregnancy, 2012, 179827. 122  87. Knipp G.T., Audus K.L. and Soares M.J. (1999) Nutrient transport across the placenta. Adv. Drug Deliv. Rev., 38, 41-58. 88. Hsiao E.Y. and Patterson P.H. (2012) Placental regulation of maternal!fetal interactions and brain development. Developmental neurobiology, 72, 1317-1326. 89. Knöfler M. and Pollheimer J. (2012) IFPA award in placentology lecture: Molecular regulation of human trophoblast invasion. Placenta, 33, S55-S62. 90. Myatt L. and Cui X. (2004) Oxidative stress in the placenta. Histochem. Cell Biol., 122, 369-382. 91. Kaufmann P. (1997) Oxygen and placental villous development: Origins of fetal hypoxia. Placenta, 18, 613-621. 92. Jansson T. and Powell T.L. (2006) Human placental transport in altered fetal growth: Does the placenta function as a nutrient sensor?–A review. Placenta, 27, 91-97. 93. Malassiné A. and Cronier L. (2002) Hormones and human trophoblast differentiation. Endocrine, 19, 3-11. 94. Goswami D., Tannetta D., Magee L., Fuchisawa A., Redman C., Sargent I. and Von Dadelszen P. (2006) Excess syncytiotrophoblast microparticle shedding is a feature of early-onset pre-eclampsia, but not normotensive intrauterine growth restriction. Placenta, 27, 56-61. 123  95. Knight M., Redman C.W., Linton E.A. and Sargent I.L. (1998) Shedding of syncytiotrophoblast microvilli into the maternal circulation in pre!eclamptic pregnancies. BJOG: An International Journal of Obstetrics & Gynaecology, 105, 632-640. 96. Wang J.X., Knottnerus A., Schuit G., Norman R.J., Chan A. and Dekker G.A. (2002) Surgically obtained sperm, and risk of gestational hypertension and pre-eclampsia. The Lancet, 359, 673-674. 97. Gudmundsson S., Dubiel M. and Sladkevicius P. (2009) Placental Morphologic and Functional Imaging in High-Risk Pregnancies. Elsevier. 98. Wang P.I., Chong S.T., Kielar A.Z., Kelly A.M., Knoepp U.D., Mazza M.B. and Goodsitt M.M. (2012) Imaging of pregnant and lactating patients: Part 1, evidence-based review and recommendations. Am. J. Roentgenol., 198, 778-784. 99. Tremblay E., Thérasse E., Thomassin-Naggara I. and Trop I. (2012) Quality initiatives: Guidelines for use of medical imaging during pregnancy and lactation. Radiographics, 32, 897-911. 100. Gudmundsson S., Dubiel M. and Sladkevicius P. (2009) Placental Morphologic and Functional Imaging in High-Risk Pregnancies. Elsevier. 101. Abramowicz J. and Sheiner E. (2007) In utero imaging of the placenta: Importance for diseases of pregnancy. Placenta, 28, S14-S22. 124  102. Prsa M., Sun L., van Amerom J., Yoo S.J., Grosse-Wortmann L., Jaeggi E., Macgowan C. and Seed M. (2014) Reference ranges of blood flow in the major vessels of the normal human fetal circulation at term by phase-contrast magnetic resonance imaging. Circ. Cardiovasc. Imaging, 7, 663-670. 103. Takalkar A.M., Khandelwal A., Lokitz S., Lilien D.L. and Stabin M.G. (2011) 18F-FDG PET in pregnancy and fetal radiation dose estimates. J. Nucl. Med., 52, 1035-1040. 104. Jones T. and Budinger T.F. (2013) The potential for low-dose functional studies in maternal-fetal medicine using PET/MR imaging. J. Nucl. Med., 54, 2016-2017. 105. Velauthar L., Plana M., Kalidindi M., Zamora J., Thilaganathan B., Illanes S., Khan K., Aquilina J. and Thangaratinam S. (2014) First!trimester uterine artery doppler and adverse pregnancy outcome: A meta!analysis involving 55 974 women. Ultrasound in Obstetrics & Gynecology, 43, 500-507. 106. Ghidini A. and Gratacos E. (2014) Can prenatal screening reduce the adverse obstetric outcomes related to abnormal placentation? Prenat. Diagn., 34, 613-617. 107. Redline R.W. (2015) Classification of placental lesions. Obstet. Gynecol., 213, S21-S28. 108. Akolekar R., Syngelaki A., Poon L., Wright D. and Nicolaides K.H. (2013) Competing risks model in early screening for preeclampsia by biophysical and biochemical markers. Fetal. Diagn. Ther., 33, 8-15. 125  109. Poon L.C., Syngelaki A., Akolekar R., Lai J. and Nicolaides K.H. (2013) Combined screening for preeclampsia and small for gestational age at 11-13 weeks. Fetal. Diagn. Ther., 33, 16-27. 110. Gagnon A., Wilson R., Audibert F., Allen V., Blight C., Brock J., Desilets V., Johnson J., Langlois S. and Summers A. (2008) Obstetrical complications associated with abnormal maternal serum markers analytes. J Obstet Gynaecol Can, 30, 918-949. 111. Chitayat D., Langlois S., Wilson R.D., Genetics Committee of the Society of Obstetricians and Gynaecologists of Canada and Prenatal Diagnosis Committee of the Canadian College of Medical Geneticists (2011) Prenatal screening for fetal aneuploidy in singleton pregnancies. J. Obstet. Gynaecol. Can., 33, 736-750. 112. Wilson K., Czerwinski J., Hoskovec J., Noblin S., Sullivan C., Harbison A., Campion M., Devary K., Devers P. and Singletary C. (2013) NSGC practice guideline: Prenatal screening and diagnostic testing options for chromosome aneuploidy. Journal of genetic counseling, 22, 4-15. 113. Audibert F., Boucoiran I., An N., Aleksandrov N., Delvin E., Bujold E. and Rey E. (2010) Screening for preeclampsia using first-trimester serum markers and uterine artery doppler in nulliparous women. Am. J. Obstet. Gynecol., 203, 383. e1-383. e8. 114. Manokhina I., Wilson S.L. and Robinson W.P. (2015). Non-invasive nucleic-acid based approaches to monitor placental health and predict pregnancy-related complications. Am J Obs Gynecol,213 S197-S206. 126  115. Kalousek D. (1985) Mosaicism confined to chorionic tissue in human gestations. First trimester fetal diagnosis, 1, 130-136. 116. Robinson W., Barrett I., Bernard L., Telenius A., Bernasconi F., Wilson R., Best R., Howard-Peebles P., Langlois S. and Kalousek D. (1997) Meiotic origin of trisomy in confined placental mosaicism is correlated with presence of fetal uniparental disomy, high levels of trisomy in trophoblast, and increased risk of fetal intrauterine growth restriction. Am. J. Hum. Genet., 60, 917. 117. Yong P.J., Langlois S., Dadelszen P.v. and Robinson W. (2006) The association between preeclampsia and placental trisomy 16 mosaicism. Prenat. Diagn., 26, 956-961. 118. Blair J., Langlois S., McFadden D. and Robinson W. (2014) Overlapping DNA methylation profile between placentas with trisomy 16 and early-onset preeclampsia. Placenta, 35, 216-222. 119. Du M.R., Zhou W.H., Dong L., Zhu X.Y., He Y.Y., Yang J.Y. and Li D.J. (2008) Cyclosporin A promotes growth and invasiveness in vitro of human first-trimester trophoblast cells via MAPK3/MAPK1-mediated AP1 and Ca2+/calcineurin/NFAT signaling pathways. Biol. Reprod., 78, 1102-1110. 120. Chakraborty C., Gleeson L.M., McKinnon T. and Lala P.K. (2002) Regulation of human trophoblast migration and invasiveness. Can. J. Physiol. Pharmacol., 80, 116-124. 121. Salonen Ros H., Lichtenstein P., Lipworth L. and Cnattingius S. (2000) Genetic effects on the liability of developing pre!eclampsia and gestational hypertension. American Journal of Medical Genetics Part A, 91, 256-260. 127  122. Nilsson E., Salonen Ros H., Cnattingius S. and Lichtenstein P. (2004) The importance of genetic and environmental effects for pre!eclampsia and gestational hypertension: A family study. BJOG: An International Journal of Obstetrics & Gynaecology, 111, 200-206. 123. Thornton J. and Macdonald A. (1999) Twin mothers, pregnancy hypertension and pre!eclampsia. BJOG: An International Journal of Obstetrics & Gynaecology, 106, 570-575. 124. Oudejans C.B., Poutsma A., Michel O.J., Thulluru H.K., Mulders J., van de Vrugt H.J., Sistermans E.A. and van Dijk M. (2016) Noncoding RNA-regulated gain-of-function of STOX2 in finnish pre-eclamptic families. Sci. Rep., 6, 32129. 125. Oudejans C.B., Michel O.J., Janssen R., Habets R., Poutsma A., Sistermans E.A., Weiss M.M., Incarnato D., Oliviero S. and Kleiverda G. (2014) Susceptibility allele-specific loss of miR-1324-mediated silencing of the INO80B chromatin-assembly complex gene in pre-eclampsia. Hum. Mol. Genet., 24, 118-127. 126. Sakowicz A., Pietrucha T., Rybak-Krzyszkowska M., Huras H., Gach A., Sakowicz B., Banaszczyk M., Grzesiak M. and Biesiada L. (2017) Double hit of NEMO gene in preeclampsia. PloS one, 12, e0180065. 127. McGinnis R., Steinthorsdottir V., Williams N.O., Thorleifsson G., Shooter S., Hjartardottir S., Bumpstead S., Stefansdottir L., Hildyard L. and Sigurdsson J.K. (2017) Variants in the fetal genome near FLT1 are associated with risk of preeclampsia. Nat. Genet.49,1255-1260. 128  128. Williams P.J. and Pipkin F.B. (2011) The genetics of pre-eclampsia and other hypertensive disorders of pregnancy. Best practice & research Clinical obstetrics & gynaecology, 25, 405-417. 129. Srinivas S.K., Morrison A.C., Andrela C.M. and Elovitz M.A. (2010) Allelic variations in angiogenic pathway genes are associated with preeclampsia. Obstet. Gynecol., 202, 445. e1-445. e11. 130. Morrison A.C., Srinivas S.K., Elovitz M.A. and Puschett J.B. (2010) Genetic variation in solute carrier genes is associated with preeclampsia. Obstet. Gynecol., 203, 491. e1-491. e13. 131. Bell M.J., Roberts J.M., Founds S.A., Jeyabalan A., Terhorst L. and Conley Y.P. (2013) Variation in endoglin pathway genes is associated with preeclampsia: A case–control candidate gene association study. BMC pregnancy and childbirth, 13, 82. 132. Haggerty C.L., Ferrell R.E., Hubel C.A., Markovic N., Harger G. and Ness R.B. (2005) Association between allelic variants in cytokine genes and preeclampsia. Obstet. Gynecol., 193, 209-215. 133. Daher S., Sass N., Oliveira L.G. and Mattar R. (2006) Cytokine genotyping in preeclampsia. American Journal of Reproductive Immunology, 55, 130-135. 134. Chappell S. and Morgan L. (2006) Searching for genetic clues to the causes of pre-eclampsia. Clin. Sci. (Lond), 110, 443-458. 129  135. Buurma A.J., Turner R.J., Driessen J.H., Mooyaart A.L., Schoones J.W., Bruijn J.A., Bloemenkamp K.W., Dekkers O.M. and Baelde H.J. (2013) Genetic variants in pre-eclampsia: A meta-analysis. Hum. Reprod. Update, 19, 289-303. 136. Colhoun H.M., McKeigue P.M. and Smith G.D. (2003) Problems of reporting genetic associations with complex outcomes. The Lancet, 361, 865-872. 137. Arngrimsson R., Siguroardottir S., Frigge M.L., Bjarnadottir R.I., Jonsson T., Stefansson H., Baldursdottir A., Einarsdottir A.S., Palsson B., Snorradottir S. et al. (1999) A genome-wide scan reveals a maternal susceptibility locus for pre-eclampsia on chromosome 2p13. Hum. Mol. Genet., 8, 1799-1805. 138. Moses E.K., Lade J.A., Guo G., Wilton A.N., Grehan M., Freed K., Borg A., Terwilliger J.D., North R. and Cooper D.W. (2000) A genome scan in families from australia and new zealand confirms the presence of a maternal susceptibility locus for pre-eclampsia, on chromosome 2. The American Journal of Human Genetics, 67, 1581-1585. 139. Laivuori H., Lahermo P., Ollikainen V., Widen E., Häivä-Mällinen L., Sundström H., Laitinen T., Kaaja R., Ylikorkala O. and Kere J. (2003) Susceptibility loci for preeclampsia on chromosomes 2p25 and 9p13 in finnish families. The American Journal of Human Genetics, 72, 168-177. 140. Harrison G.A., Humphrey K.E., Jones N., Badenhop R., Guo G., Elakis G., Kaye J.A., Turner R.J., Grehan M., Wilton A.N. et al. (1997) A genomewide linkage study of 130  preeclampsia/eclampsia reveals evidence for a candidate region on 4q. Am. J. Hum. Genet., 60, 1158-1167. 141. Lachmeijer A.M., Arngrímsson R., Bastiaans E.J., Frigge M.L., Pals G., Sigurdardóttir S., Stéfansson H., Pálsson B., Nicolae D. and Kong A. (2001) A genome-wide scan for preeclampsia in the netherlands. European journal of human genetics: EJHG, 9, 758. 142. Morgan L., McGinnis R., Steinthorsdottir V., Svyatova G., Zakhidova N., Lee W.K., Iversen A.C., Magnus P., Walker J., Casas J.P. et al. (2014) InterPregGen: Genetic studies of pre-eclampsia in three continents. Nor. Epidemiol., 24, 141-146. 143. Zhao L., Triche E.W., Walsh K.M., Bracken M.B., Saftlas A.F., Hoh J. and Dewan A.T. (2012) Genome-wide association study identifies a maternal copy-number deletion in PSG11 enriched among preeclampsia patients. BMC pregnancy and childbirth, 12, 61. 144. Johnson M.P., Brennecke S.P., East C.E., Göring H.H., Kent Jr J.W., Dyer T.D., Said J.M., Roten L.T., Iversen A. and Abraham L.J. (2012) Genome-wide association scan identifies a risk locus for preeclampsia on 2q14, near the inhibin, beta B gene. PloS one, 7, e33666. 145. Lee E.S., Oh M.J., Jung J.W., Lim J.E., Seol H.J., Lee K.J. and Kim H.J. (2007) The levels of circulating vascular endothelial growth factor and soluble flt-1 in pregnancies complicated by preeclampsia. J. Korean Med. Sci., 22, 94-98. 146. Enquobahrie D.A., Meller M., Rice K., Psaty B.M., Siscovick D.S. and Williams M.A. (2008) Differential placental gene expression in preeclampsia. Am. J. Obstet. Gynecol., 199, 566. e1-566. e11. 131  147. Sitras V., Paulssen R., Grønaas H., Leirvik J., Hanssen T., Vårtun Å and Acharya G. (2009) Differential placental gene expression in severe preeclampsia. Placenta, 30, 424-433. 148. Hewitt G., Jurk D., Marques F.D., Correia-Melo C., Hardy T., Gackowska A., Anderson R., Taschuk M., Mann J. and Passos J.F. (2012) Telomeres are favoured targets of a persistent DNA damage response in ageing and stress-induced senescence. Nature communications, 3, 708. 149. Davy P., Nagata M., Bullard P., Fogelson N. and Allsopp R. (2009) Fetal growth restriction is associated with accelerated telomere shortening and increased expression of cell senescence markers in the placenta. Placenta, 30, 539-542. 150. Biron-Shental T., Sukenik-Halevy R., Sharon Y., Goldberg-Bittman L., Kidron D., Fejgin M.D. and Amiel A. (2010) Short telomeres may play a role in placental dysfunction in preeclampsia and intrauterine growth restriction. Am. J. Obstet. Gynecol., 202, 381. e1-381. e7. 151. Biron-Shental T., Sukenik Halevy R., Goldberg-Bittman L., Kidron D., Fejgin M.D. and Amiel A. (2010) Telomeres are shorter in placental trophoblasts of pregnancies complicated with intrauterine growth restriction (IUGR). Early Hum. Dev., 86, 451-456. 152. Toutain J., Prochazkova-Carlotti M., Cappellen D., Jarne A., Chevret E., Ferrer J., Idrissi Y., Pelluard F., Carles D. and Maugey-Laulon B. (2013) Reduced placental telomere length during pregnancies complicated by intrauterine growth restriction. PloS one, 8, e54013. 153. Gielen M., Hageman G., Pachen D., Derom C., Vlietinck R. and Zeegers M. (2014) Placental telomere length decreases with gestational age and is influenced by parity: A study of third trimester live-born twins. Placenta, 35, 791-796. 132  154. Founds S.A., Conley Y.P., Lyons-Weiler J.F., Jeyabalan A., Hogge W.A. and Conrad K.P. (2009) Altered global gene expression in first trimester placentas of women destined to develop preeclampsia. Placenta, 30, 15-24. 155. Dunk C., Roggensack A., Cox B., Perkins J., Åsenius F., Keating S., Weksberg R., Kingdom J. and Adamson S. (2012) A distinct microvascular endothelial gene expression profile in severe IUGR placentas. Placenta, 33, 285-293. 156. Guo L., Choufani S., Ferreira J., Smith A., Chitayat D., Shuman C., Uxa R., Keating S., Kingdom J. and Weksberg R. (2008) Altered gene expression and methylation of the human chromosome 11 imprinted region in small for gestational age (SGA) placentae. Dev. Biol., 320, 79-91. 157. McCarthy C., Cotter F.E., McElwaine S., Twomey A., Mooney E.E., Ryan F. and Vaughan J. (2007) Altered gene expression patterns in intrauterine growth restriction: Potential role of hypoxia. Am. J. Obstet. Gynecol., 196, 70. e1-70. e6. 158. Enquobahrie D.A., Meller M., Rice K., Psaty B.M., Siscovick D.S. and Williams M.A. (2008) Differential placental gene expression in preeclampsia. Am. J. Obstet. Gynecol., 199, 566. e1-566. e11. 159. Nevo O., Many A., Xu J., Kingdom J., Piccoli E., Zamudio S., Post M., Bocking A., Todros T. and Caniggia I. (2008) Placental expression of soluble fms-like tyrosine kinase 1 is increased in singletons and twin pregnancies with intrauterine growth restriction. Journal of Clinical Endocrinology & Metabolism, 93, 285-292. 133  160. Regnault T., Orbus R., De Vrijer B., Davidsen M., Galan H., Wilkening R. and Anthony R. (2002) Placental expression of VEGF, PlGF and their receptors in a model of placental insufficiency—intrauterine growth restriction (PI-IUGR). Placenta, 23, 132-144. 161. Winn V.D., Gormley M., Paquet A.C., Kjaer-Sorensen K., Kramer A., Rumer K.K., Haimov-Kochman R., Yeh R., Overgaard M.T. and Varki A. (2009) Severe preeclampsia-related changes in gene expression at the maternal-fetal interface include sialic acid-binding immunoglobulin-like lectin-6 and pappalysin-2. Endocrinology, 150, 452-462. 162. Nevo O., Many A., Xu J., Kingdom J., Piccoli E., Zamudio S., Post M., Bocking A., Todros T. and Caniggia I. (2008) Placental expression of soluble fms-like tyrosine kinase 1 is increased in singletons and twin pregnancies with intrauterine growth restriction. Journal of Clinical Endocrinology & Metabolism, 93, 285-292. 163. Enquobahrie D.A., Abetew D.F., Sorensen T.K., Willoughby D., Chidambaram K. and Williams M.A. (2011) Placental microRNA expression in pregnancies complicated by preeclampsia. Obstet. Gynecol., 204, 178. e12-178. e21. 164. Avila L., Yuen R., Diego-Alvarez D., Peñaherrera M., Jiang R. and Robinson W. (2010) Evaluating DNA methylation and gene expression variability in the human term placenta. Placenta, 31, 1070-1077. 165. Roland L., Beauchemin D., Acteau G., Fradette C., St-Pierre I. and Bilodeau J. (2010) Effects of labor on placental expression of superoxide dismutases in preeclampsia. Placenta, 31, 392-400. 134  166. Lee K., Shim S., Kang K., Kang J., Park D., Kim S., Farina A., Shim S. and Cha D. (2010) Global gene expression changes induced in the human placenta during labor. Placenta, 31, 698-704. 167. Tsochandaridis M., Nasca L., Toga C. and Levy-Mozziconacci A. (2015) Circulating microRNAs as clinical biomarkers in the predictions of pregnancy complications. BioMed research international, 2015. 168. Anton L., Olarerin-George A.O., Schwartz N., Srinivas S., Bastek J., Hogenesch J.B. and Elovitz M.A. (2013) miR-210 inhibits trophoblast invasion and is a serum biomarker for preeclampsia. The American journal of pathology, 183, 1437-1445. 169. Russo V.E., Martienssen R.A. and Riggs A.D. (1996) Epigenetic Mechanisms of Gene Regulation. Cold Spring Harbor Laboratory Press. 170. Schübeler D. (2015) Function and information content of DNA methylation. Nature, 517, 321-326. 171. Cooper D.N. and Krawczak M. (1989) Cytosine methylation and the fate of CpG dinucleotides in vertebrate genomes. Hum. Genet., 83, 181-188. 172. Fatemi M., Hermann A., Pradhan S. and Jeltsch A. (2001) The activity of the murine DNA methyltransferase Dnmt1 is controlled by interaction of the catalytic domain with the N-terminal part of the enzyme leading to an allosteric activation of the enzyme after binding to methylated DNA. J. Mol. Biol., 309, 1189-1199. 135  173. Okano M., Bell D.W., Haber D.A. and Li E. (1999) DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell, 99, 247-257. 174. Deaton A.M. and Bird A. (2011) CpG islands and the regulation of transcription. Genes Dev., 25, 1010-1022. 175. Razin A. and Cedar H. (1991) DNA methylation and gene expression. Microbiol. Rev., 55, 451-458. 176. Klose R.J. and Bird A.P. (2006) Genomic DNA methylation: The mark and its mediators. Trends Biochem. Sci., 31, 89-97. 177. Jurkowska R.Z., Jurkowski T.P. and Jeltsch A. (2011) Structure and function of mammalian DNA methyltransferases. Chembiochem, 12, 206-222. 178. Jjingo D., Conley A.B., Soojin V.Y., Lunyak V.V. and Jordan I.K. (2012) On the presence and role of human gene-body DNA methylation. Oncotarget, 3, 462. 179. Zhang F.F., Cardarelli R., Carroll J., Fulda K.G., Kaur M., Gonzalez K., Vishwanatha J.K., Santella R.M. and Morabia A. (2011) Significant differences in global genomic DNA methylation by gender and race/ethnicity in peripheral blood. Epigenetics, 6, 623-629. 180. Singmann P., Shem-Tov D., Wahl S., Grallert H., Fiorito G., Shin S., Schramm K., Wolf P., Kunze S. and Baran Y. (2015) Characterization of whole-genome autosomal differences of DNA methylation between men and women. Epigenetics & chromatin, 8, 43. 136  181. Lee J.T. and Bartolomei M.S. (2013) X-inactivation, imprinting, and long noncoding RNAs in health and disease. Cell, 152, 1308-1323. 182. Gabory A., Attig L. and Junien C. (2009) Sexual dimorphism in environmental epigenetic programming. Mol. Cell. Endocrinol., 304, 8-18. 183. Connor C.M. and Akbarian S. (2008) DNA methylation changes in schizophrenia and bipolar disorder. Epigenetics, 3, 55-58. 184. Gabory A., Roseboom T.J., Moore T., Moore L.G. and Junien C. (2013) Placental contribution to the origins of sexual dimorphism in health and diseases: Sex chromosomes and epigenetics. Biology of sex differences, 4, 5. 185. Kundakovic M., Gudsnuk K., Franks B., Madrid J., Miller R.L., Perera F.P. and Champagne F.A. (2013) Sex-specific epigenetic disruption and behavioral changes following low-dose in utero bisphenol A exposure. Proc. Natl. Acad. Sci. U. S. A., 110, 9956-9961. 186. Lévesque M.L., Casey K.F., Szyf M., Ismaylova E., Ly V., Verner M., Suderman M., Brendgen M., Vitaro F. and Dionne G. (2014) Genome-wide DNA methylation variability in adolescent monozygotic twins followed since birth. Epigenetics, 9, 1410-1422. 187. Ollikainen M., Ismail K., Gervin K., Kyllönen A., Hakkarainen A., Lundbom J., Järvinen E.A., Harris J.R., Lundbom N. and Rissanen A. (2015) Genome-wide blood DNA methylation alterations at regulatory elements and heterochromatic regions in monozygotic twins discordant for obesity and liver fat. Clinical epigenetics, 7, 39. 137  188. Chiarella J., Tremblay R.E., Szyf M., Provençal N. and Booij L. (2015) Impact of early environment on children's mental health: Lessons from DNA methylation studies with monozygotic twins. Twin Research and Human Genetics, 18, 623-634. 189. Heindel J.J. and Vandenberg L.N. (2015) Developmental origins of health and disease: A paradigm for understanding disease cause and prevention. Curr. Opin. Pediatr., 27, 248-253. 190. Gluckman P.D., Buklijas T. and Hanson M.A. (2015) The developmental origins of health and disease (DOHaD) concept: Past, present, and future. The epigenome and developmental origins of health and disease.Academic, London, , 1-13. 191. Hanson M.A. and Gluckman P.D. (2014) Early developmental conditioning of later health and disease: Physiology or pathophysiology? Physiol. Rev., 94, 1027-1076. 192. De Souza R.A., Islam S.A., McEwen L.M., Mathelier A., Hill A., Mah S.M., Wasserman W.W., Kobor M.S. and Leavitt B.R. (2016) DNA methylation profiling in human huntington's disease brain. Hum. Mol. Genet., 25, 2013-2030. 193. Coppieters N., Dieriks B.V., Lill C., Faull R.L., Curtis M.A. and Dragunow M. (2014) Global changes in DNA methylation and hydroxymethylation in alzheimer's disease human brain. Neurobiol. Aging, 35, 1334-1344. 194. De Jager P.L., Srivastava G., Lunnon K., Burgess J., Schalkwyk L.C., Yu L., Eaton M.L., Keenan B.T., Ernst J. and McCabe C. (2014) Alzheimer's disease: Early alterations in brain DNA methylation at ANK1, BIN1, RHBDF2 and other loci. Nat. Neurosci., 17, 1156-1163. 138  195. Pries L., Gülöksüz S. and Kenis G. (2017) DNA Methylation in Schizophrenia. In Neuroepigenomics in Aging and Disease. Springer, pp. 211-236. 196. Janzen C., Lei M.Y., Cho J., Sullivan P., Shin B. and Devaskar S.U. (2013) Placental glucose transporter 3 (GLUT3) is up-regulated in human pregnancies complicated by late-onset intrauterine growth restriction. Placenta, 34, 1072-1078. 197. Trerotola M., Relli V., Simeone P. and Alberti S. (2015) Epigenetic inheritance and the missing heritability. Human genomics, 9, 17. 198. Koch L. (2014) Epigenetics: An epigenetic twist on the missing heritability of complex traits. Nature Reviews Genetics, 15, 218-218. 199. Lasseigne B.N., Burwell T.C., Patil M.A., Absher D.M., Brooks J.D. and Myers R.M. (2014) DNA methylation profiling reveals novel diagnostic biomarkers in renal cell carcinoma. BMC medicine, 12, 235. 200. Heyn H., Carmona F.J., Gomez A., Ferreira H.J., Bell J.T., Sayols S., Ward K., Stefansson O.A., Moran S., Sandoval J. et al. (2013) DNA methylation profiling in breast cancer discordant identical twins identifies DOK7 as novel epigenetic biomarker. Carcinogenesis, 34, 102-108. 201. Wu P., Farrell W.E., Haworth K.E., Emes R.D., Kitchen M.O., Glossop J.R., Hanna F.W. and Fryer A.A. (2016) Maternal genome-wide DNA methylation profiling in gestational diabetes shows distinctive disease-associated changes relative to matched healthy pregnancies. Epigenetics, , 00-00. 139  202. Galanopoulos M., Tsoukalas N., Papanikolaou I.S., Tolia M., Gazouli M. and Mantzaris G.J. (2017) Abnormal DNA methylation as a cell-free circulating DNA biomarker for colorectal cancer detection: A review of literature. World Journal of Gastrointestinal Oncology, 9, 142. 203. Guo S., Diep D., Plongthongkum N., Fung H., Zhang K. and Zhang K. (2017) Identification of methylation haplotype blocks aids in deconvolution of heterogeneous tissue samples and tumor tissue-of-origin mapping from plasma DNA. Nat. Genet., 49, 635-642. 204. Weiss G.J., Beck J., Braun D.P., Bornemann-Kolatzki K., Barilla H., Cubello R., Quan W., Sangal A., Khemka V., Waypa J. et al. (2017) Tumor cell-free DNA copy number instability predicts therapeutic response to immunotherapy. Clin. Cancer Res.,clinres-0231 . 205. Tanić M. and Beck S. (2017) Epigenome-wide association studies for cancer biomarker discovery in circulating cell-free DNA: Technical advances and challenges. Curr. Opin. Genet. Dev., 42, 48-55. 206. Bedin C., Enzo M.V., Del Bianco P., Pucciarelli S., Nitti D. and Agostini M. (2017) Diagnostic and prognostic role of cell!free DNA testing for colorectal cancer patients. International Journal of Cancer,140, 1888-1898. 207. Li H., Guo Z., Guo Y., Li M., Yan H., Cheng J., Wang C. and Hong G. (2016) Common DNA methylation alterations of alzheimer's disease and aging in peripheral whole blood. Oncotarget, 7, 19089-19098. 208. Sanchez-Mut J., Heyn H., Vidal E., Moran S., Sayols S., Delgado-Morales R., Schultz M., Ansoleaga B., Garcia-Esparcia P. and Pons-Espinal M. (2016) Human DNA methylomes of 140  neurodegenerative diseases show common epigenomic patterns. Translational psychiatry, 6, e718. 209. Bhargava A., Kumar Khare N., Bunkar N., Chaudhury K., Chand Pandey K., K Jain S. and K Mishra P. (2017) Cell-free circulating epigenomic signatures: Non-invasive biomarker for cardiovascular and other age-related chronic diseases. Curr. Pharm. Des., 23, 1175-1187. 210. Varley K.E., Gertz J., Bowling K.M., Parker S.L., Reddy T.E., Pauli-Behn F., Cross M.K., Williams B.A., Stamatoyannopoulos J.A., Crawford G.E. et al. (2013) Dynamic DNA methylation across diverse human cell lines and tissues. Genome Res., 23, 555-567. 211. Jaffe A.E. and Irizarry R.A. (2014) Accounting for cellular heterogeneity is critical in epigenome-wide association studies. Genome Biol., 15, R31. 212. Suelves M., Carrio E., Nunez-Alvarez Y. and Peinado M.A. (2016) DNA methylation dynamics in cellular commitment and differentiation. Brief Funct. Genomics, 15, 443-453. 213. Houseman E.A., Accomando W.P., Koestler D.C., Christensen B.C., Marsit C.J., Nelson H.H., Wiencke J.K. and Kelsey K.T. (2012) DNA methylation arrays as surrogate measures of cell mixture distribution. BMC Bioinformatics, 13, 86. 214. Montaño C.M., Irizarry R.A., Kaufmann W.E., Talbot K., Gur R.E., Feinberg A.P. and Taub M.A. (2013) Measuring cell-type specific differential methylation in human brain tissue. Genome Biol., 14, R94. 141  215. Houseman E.A., Molitor J. and Marsit C.J. (2014) Reference-free cell mixture adjustments in analysis of DNA methylation data. Bioinformatics, 30, 1431-1439. 216. Jones M.J., Goodman S.J. and Kobor M.S. (2015) DNA methylation and healthy human aging. Aging cell, 14, 924-932. 217. Day K., Waite L.L., Thalacker-Mercer A., West A., Bamman M.M., Brooks J.D., Myers R.M. and Absher D. (2013) Differential DNA methylation with age displays both common and dynamic features across human tissues that are influenced by CpG landscape. Genome Biol., 14, R102. 218. Horvath S. (2013) DNA methylation age of human tissues and cell types. Genome Biol., 14, 3156. 219. Bocklandt S., Lin W., Sehl M.E., Sánchez F.J., Sinsheimer J.S., Horvath S. and Vilain E. (2011) Epigenetic predictor of age. PloS one, 6, e14821. 220. Hannum G., Guinney J., Zhao L., Zhang L., Hughes G., Sadda S., Klotzle B., Bibikova M., Fan J. and Gao Y. (2013) Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol. Cell, 49, 359-367. 221. Mayne B.T., Leemaqz S.Y., Smith A.K., Breen J., Roberts C.T. and Bianco-Miotto T. (2017) Accelerated placental aging in early onset preeclampsia pregnancies identified by DNA methylation. Epigenomics, 9, 279-289. 142  222. Chen B.H., Marioni R.E., Colicino E., Peters M.J., Ward-Caviness C.K., Tsai P.C., Roetker N.S., Just A.C., Demerath E.W., Guan W. et al. (2016) DNA methylation-based measures of biological age: Meta-analysis predicting time to death. Aging (Albany NY), 8, 1844-1865. 223. Perna L., Zhang Y., Mons U., Holleczek B., Saum K. and Brenner H. (2016) Epigenetic age acceleration predicts cancer, cardiovascular, and all-cause mortality in a german case cohort. Clinical epigenetics, 8, 64. 224. Levine M.E., Hosgood H.D., Chen B., Absher D., Assimes T. and Horvath S. (2015) DNA methylation age of blood predicts future onset of lung cancer in the women's health initiative. Aging (Albany NY), 7, 690-700. 225. Zheng Y., Joyce B.T., Colicino E., Liu L., Zhang W., Dai Q., Shrubsole M.J., Kibbe W.A., Gao T. and Zhang Z. (2016) Blood epigenetic age may predict cancer incidence and mortality. EBioMedicine, 5, 68-73. 226. Zannas A.S., Arloth J., Carrillo-Roa T., Iurato S., Röh S., Ressler K.J., Nemeroff C.B., Smith A.K., Bradley B. and Heim C. (2015) Lifetime stress accelerates epigenetic aging in an urban, african american cohort: Relevance of glucocorticoid signaling. Genome Biol., 16, 266. 227. Horvath S., Langfelder P., Kwak S., Aaronson J., Rosinski J., Vogt T.F., Eszes M., Faull R.L., Curtis M.A., Waldvogel H.J. et al. (2016) Huntington's disease accelerates epigenetic aging of human brain and disrupts DNA methylation levels. Aging (Albany NY), 8, 1485-1512. 143  228. Horvath S., Erhart W., Brosch M., Ammerpohl O., von Schonfels W., Ahrens M., Heits N., Bell J.T., Tsai P.C., Spector T.D. et al. (2014) Obesity accelerates epigenetic aging of human liver. Proc. Natl. Acad. Sci. U. S. A., 111, 15538-15543. 229. Winn V.D., Gormley M. and Fisher S.J. (2011) The impact of preeclampsia on gene expression at the maternal–fetal interface. Pregnancy Hypertension: An International Journal of Women's Cardiovascular Health, 1, 100-108. 230. Novakovic B., Yuen R., Gordon L., Penaherrera M., Sharkey A., Moffett A., Craig J., Robinson W. and Saffery R. (2011) Evidence for widespread changes in promoter methylation profile in human placenta in response to increasing gestational age and environmental/stochastic factors. BMC Genomics, 12, 529. 231. Reik W. (2007) Stability and flexibility of epigenetic gene regulation in mammalian development. Nature, 447, 425. 232. Reik W., Dean W. and Walter J. (2001) Epigenetic reprogramming in mammalian development. Science, 293, 1089-1093. 233. Schroeder D.I. and LaSalle J.M. (2013) How has the study of the human placenta aided our understanding of partially methylated genes? . 234. Fowden A.L., Sibley C., Reik W. and Constancia M. (2006) Imprinted genes, placental development and fetal growth. Horm. Res., 65 Suppl 3, 50-58. 144  235. Du P., Zhang X., Huang C., Jafari N., Kibbe W.A., Hou L. and Lin S.M. (2010) Comparison of beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC Bioinformatics, 11, 587. 236. Ehrlich M., Gama-Sosa M.A., Huang L.H., Midgett R.M., Kuo K.C., McCune R.A. and Gehrke C. (1982) Amount and distribution of 5-methylcytosine in human DNA from different types of tissues of cells. Nucleic Acids Res., 10, 2709-2721. 237. Price E.M., Cotton A.M., Peñaherrera M.S., McFadden D.E., Kobor M.S. and Robinson W. (2012) Different measures of “genome-wide” DNA methylation exhibit unique properties in placental and somatic tissues. Epigenetics, 7, 652-663. 238. Avila L., Yuen R., Diego-Alvarez D., Peñaherrera M., Jiang R. and Robinson W. (2010) Evaluating DNA methylation and gene expression variability in the human term placenta. Placenta, 31, 1070-1077. 239. Kim J., Cheong H.S., Lee D.S., Shin H.D. and Kim Y.N. Genome-wide DNA methylation profiles of maternal peripheral blood and placentas: Potential risk factors for preeclampsia and validation of GRK5. Genes & Genomics,39,197-206. 240. Yeung K.R., Chiu C.L., Pidsley R., Makris A., Hennessy A. and Lind J.M. (2016) DNA methylation profiles in preeclampsia and healthy control placentas. Am. J. Physiol. Heart Circ. Physiol., 310, H1295-303. 241. Hillman S.L., Finer S., Smart M.C., Mathews C., Lowe R., Rakyan V.K., Hitman G.A. and Williams D.J. (2015) Novel DNA methylation profiles associated with key gene regulation and 145  transcription pathways in blood and placenta of growth-restricted neonates. Epigenetics, 10, 50-61. 242. Zhu L., Lv R., Kong L., Cheng H., Lan F. and Li X. (2015) Genome-wide mapping of 5mC and 5hmC identified differentially modified genomic regions in late-onset severe preeclampsia: A pilot study. PloS one, 10, e0134119. 243. Anton L., Brown A.G., Bartolomei M.S. and Elovitz M.A. (2014) Differential methylation of genes associated with cell adhesion in preeclamptic placentas. PLoS One, 9, e100148. 244. Ching T., Song M.A., Tiirikainen M., Molnar J., Berry M., Towner D. and Garmire L.X. (2014) Genome-wide hypermethylation coupled with promoter hypomethylation in the chorioamniotic membranes of early onset pre-eclampsia. Mol. Hum. Reprod., 20, 885-904. 245. Anderson C.M., Ralph J.L., Wright M.L., Linggi B. and Ohm J.E. (2014) DNA methylation as a biomarker for preeclampsia. Biol. Res. Nurs., 16, 409-420. 246. Chu T., Bunce K., Shaw P., Shridhar V., Althouse A., Hubel C. and Peters D. (2014) Comprehensive analysis of preeclampsia-associated DNA methylation in the placenta. PloS one, 9, e107318. 247. Nomura Y., Lambertini L., Rialdi A., Lee M., Mystal E.Y., Grabie M., Manaster I., Huynh N., Finik J. and Davey M. (2014) Global methylation in the placenta and umbilical cord blood from pregnancies with maternal gestational diabetes, preeclampsia, and obesity. Reproductive sciences, 21, 131-137. 146  248. Yan Y., Yi P., Zheng Y., Yu L., Han J., Han X. and Li L. (2013) Screening for preeclampsia pathogenesis related genes. Eur. Rev. Med. Pharmacol. Sci., 17, 3083-3094. 249. Liu L., Zhang X., Rong C., Rui C., Ji H., Qian Y.J., Jia R. and Sun L. (2014) Distinct DNA methylomes of human placentas between pre-eclampsia and gestational diabetes mellitus. Cell. Physiol. Biochem., 34, 1877-1889. 250. Rahat B., Najar R.A., Hamid A., Bagga R. and Kaur J. (2017) The role of aberrant methylation of trophoblastic stem cell origin in the pathogenesis and diagnosis of placental disorders. Prenat. Diagn., 37, 133-143. 251. Majchrzak-Celińska A., Kosicka K., Paczkowska J., Główka F.K., Bręborowicz G.H., Krzyścin M., Siemiątkowska A., Szaumkessel M. and Baer-Dubowska W. (2017) HSD11B2, RUNX3, and LINE-1 methylation in placental DNA of hypertensive disorders of pregnancy patients. Reproductive Sciences, , 1933719117692043. 252. Xiao X., Tao X., Wang Y., Zhu L., Ye Y., Liu H., Zhou Q., Li X. and Xiong Y. (2017) Hypomethylation of tissue factor pathway inhibitor 2 in human placenta of preeclampsia. Thromb. Res., 152, 7-13. 253. Ye W., Shen L., Xiong Y., Zhou Y., Gu H. and Yang Z. (2016) Preeclampsia is associated with decreased methylation of the GNA12 promoter. Ann. Hum. Genet., 80, 7-10. 254. Lin L., Yu Y., Zhang Z. and Yang Y. (2016) Significant hypomethylation of TNFAIP8 and increased expression in the placenta and peripheral blood cells from early-onset preeclamptic 147  patients. INTERNATIONAL JOURNAL OF CLINICAL AND EXPERIMENTAL MEDICINE, 9, 10384. 255. Anderson C.M., Ralph J.L., Johnson L., Scheett A., Wright M.L., Taylor J.Y., Ohm J.E. and Uthus E. (2015) First trimester vitamin D status and placental epigenomics in preeclampsia among northern plains primiparas. Life Sci., 129, 10-15. 256. Tang Y., Liu H., Li H., Peng T., Gu W. and Li X. (2015) Hypermethylation of the HLA-G promoter is associated with preeclampsia. Mol. Hum. Reprod., 21, 736-744. 257. Xiao X., Zhao Y., Jin R., Chen J., Wang X., Baccarelli A. and Zhang Y. (2015) Fetal growth restriction and methylation of growth-related genes in the placenta.Epigenomics, 8,33-42.  258. Lu L., Hou Z., Li L., Yang Y., Wang X., Zhang B., Ren M., Zhao D., Miao Z. and Yu L. (2014) Methylation pattern of H19 exon 1 is closely related to preeclampsia and trophoblast abnormalities. Int. J. Mol. Med., 34, 765-771. 259. Liu Q., Qiao F., Shi X., Liu H., Gong X. and Wu Y. (2014) Promoter hypomethylation and increased maspin expression in preeclamptic placentas in a chinese population. Placenta, 35, 876-882. 260. Zhao Y., Gong X., Chen L., Li L., Liang Y., Chen S. and Zhang Y. (2014) Site-specific methylation of placental HSD11B2 gene promoter is related to intrauterine growth restriction. European Journal of Human Genetics, 22, 734-740. 148  261. Sundrani D.P., Reddy U.S., Joshi A.A., Mehendale S.S., Chavan-Gautam P.M., Hardikar A.A., Chandak G.R. and Joshi S.R. (2013) Differential placental methylation and expression of VEGF, FLT-1 and KDR genes in human term and preterm preeclampsia. Clin. Epigenetics, 5, 6-7083-5-6. 262. Xiang Y., Cheng Y., Li X., Li Q., Xu J., Zhang J., Liu Y., Xing Q., Wang L. and He L. (2013) Up-regulated expression and aberrant DNA methylation of LEP and SH3PXD2A in pre-eclampsia. PloS one, 8, e59753. 263. Xiang Y., Zhang X., Li Q., Xu J., Zhou X., Wang T., Xing Q., Liu Y., Wang L., He L. et al. (2013) Promoter hypomethylation of TIMP3 is associated with pre-eclampsia in a chinese population. Mol. Hum. Reprod., 19, 153-159. 264. Koukoura O., Sifakis S., Soufla G., Zaravinos A., Apostolidou S., Jones A., Widschwendter M. and Spandidos D.A. (2011) Loss of imprinting and aberrant methylation of IGF2 in placentas from pregnancies complicated with fetal growth restriction. Int. J. Mol. Med., 28, 481. 265. Gao W., Li D., Xiao Z., Liao Q., Yang H., Li Y., Ji L. and Wang Y. (2011) Detection of global DNA methylation and paternally imprinted H19 gene methylation in preeclamptic placentas. Hypertension Research, 34, 655-661. 266. Lambertini L., Lee T., Chan W., Lee M., Diplas A., Wetmur J. and Chen J. (2011) Differential methylation of imprinted genes in growth-restricted placentas. Reproductive sciences, 18, 1111-1117. 149  267. Bourque D., Avila L., Penaherrera M., Von Dadelszen P. and Robinson W. (2010) Decreased placental methylation at the H19/IGF2 imprinting control region is associated with normotensive intrauterine growth restriction but not preeclampsia. Placenta, 31, 197-202. 268. Wang Z., Lu S., Liu C., Zhao B., Pei K., Tian L. and Ma X. (2010) Expressional and epigenetic alterations of placental matrix metalloproteinase 9 in preeclampsia. Gynecological Endocrinology, 26, 96-102. 269. Chelbi S.T., Mondon F., Jammes H., Buffat C., Mignot T.M., Tost J., Busato F., Gut I., Rebourcet R., Laissue P. et al. (2007) Expressional and epigenetic alterations of placental serine protease inhibitors: SERPINA3 is a potential marker of preeclampsia. Hypertension, 49, 76-83. 270. Iglesias-Platas I., Trujillo A.M., Court F. and Monk D. (2015) Distinct promoter methylation and isoform-specific expression of RASFF1A in placental biopsies from complicated pregnancies. Placenta, 36, 397-402. 271. Bibikova M., Barnes B., Tsan C., Ho V., Klotzle B., Le J.M., Delano D., Zhang L., Schroth G.P. and Gunderson K.L. (2011) High density DNA methylation array with single CpG site resolution. Genomics, 98, 288-295. 272. Laird P.W. (2010) Principles and challenges of genome-wide DNA methylation analysis. Nature Reviews Genetics, 11, 191-203. 273. Bock C., Tomazou E.M., Brinkman A.B., Müller F., Simmer F., Gu H., Jäger N., Gnirke A., Stunnenberg H.G. and Meissner A. (2010) Quantitative comparison of genome-wide DNA methylation mapping technologies. Nat. Biotechnol., 28, 1106-1114. 150  274. Hansen K.D., Langmead B. and Irizarry R.A. (2012) BSmooth: From whole genome bisulfite sequencing reads to differentially methylated regions. Genome Biol., 13, R83. 275. Busato F. and Tost J. (2015) DNA methylation analysis by pyrosequencing. Encyclopedia of Metagenomics: Genes, Genomes and Metagenomes: Basics, Methods, Databases and Tools, , 132-141. 276. Tost J. and Gut I.G. (2007) DNA methylation analysis by pyrosequencing. Nature protocols, 2, 2265-2275. 277. Novakovic B., Yuen R.K., Gordon L., Penaherrera M.S., Sharkey A., Moffett A., Craig J.M., Robinson W.P. and Saffery R. (2011) Evidence for widespread changes in promoter methylation profile in human placenta in response to increasing gestational age and environmental/stochastic factors. BMC Genomics, 12, 529. 278. Kramer M.S., Platt R.W., Wen S.W., Joseph K.S., Allen A., Abrahamowicz M., Blondel B., Breart G. and Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System (2001) A new and improved population-based canadian reference for birth weight for gestational age. Pediatrics, 108, E35. 279. Magee L.A., Pels A., Helewa M., Rey E., von Dadelszen P., Audibert F., Bujold E., Côté A., Douglas M.J. and Eastabrook G. (2014) Diagnosis, evaluation, and management of the hypertensive disorders of pregnancy: Executive summary. Journal of Obstetrics and Gynaecology Canada, 36, 416-438. 151  280. Von Dadelszen P., Magee L.A. and Roberts J.M. (2003) Subclassification of preeclampsia. Hypertension in pregnancy, 22, 143-148. 281. Pidsley R., Zotenko E., Peters T.J., Lawrence M.G., Risbridger G.P., Molloy P., Van Djik S., Muhlhausler B., Stirzaker C. and Clark S.J. (2016) Critical evaluation of the illumina MethylationEPIC BeadChip microarray for whole-genome DNA methylation profiling. Genome Biol., 17, 208. 282. Fortin J., Labbe A., Lemire M., Zanke B.W., Hudson T.J., Fertig E.J., Greenwood C.M. and Hansen K.D. (2014) Functional normalization of 450k methylation array data improves replication in large cancer studies. Genome Biol., 15, 1-17. 283. Maksimovic J., Gordon L. and Oshlack A. (2012) SWAN: Subset-quantile within array normalization for illumina infinium HumanMethylation450 BeadChips. Genome Biol., 13, R44. 284. Touleimat N. and Tost J. (2012) Complete pipeline for infinium® human methylation 450K BeadChip data processing using subset quantile normalization for accurate DNA methylation estimation. Epigenomics, 4, 325-341. 285. Teschendorff A.E., Marabita F., Lechner M., Bartlett T., Tegner J., Gomez-Cabrero D. and Beck S. (2013) A beta-mixture quantile normalization method for correcting probe design bias in illumina infinium 450 k DNA methylation data. Bioinformatics, 29, 189-196. 286. Pidsley R., Wong C.C., Volta M., Lunnon K., Mill J. and Schalkwyk L.C. (2013) A data-driven approach to preprocessing illumina 450K methylation array data. BMC Genomics, 14, 293. 152  287. Yousefi P., Huen K., Schall R.A., Decker A., Elboudwarej E., Quach H., Barcellos L. and Holland N. (2013) Considerations for normalization of DNA methylation data by illumina 450K BeadChip assay in population studies. Epigenetics, 8, 1141-1152. 288. Marabita F., Almgren M., Lindholm M.E., Ruhrmann S., Fagerström-Billai F., Jagodic M., Sundberg C.J., Ekström T.J., Teschendorff A.E. and Tegnér J. (2013) An evaluation of analysis pipelines for DNA methylation profiling using the illumina HumanMethylation450 BeadChip platform. Epigenetics, 8, 333-346. 289. Johnson W.E., Li C. and Rabinovic A. (2007) Adjusting batch effects in microarray expression data using empirical bayes methods. Biostatistics, 8, 118-127. 290. Buhule O.D., Minster R.L., Hawley N.L., Medvedovic M., Sun G., Viali S., Deka R., McGarvey S.T. and Weeks D.E. (2014) Stratified randomization controls better for batch effects in 450K methylation analysis: A cautionary tale. Frontiers in genetics, 5, 354. 291. Price E.M., Cotton A.M., Lam L.L., Farré P., Emberly E., Brown C.J., Robinson W.P. and Kobor M.S. (2013) Additional annotation enhances potential for biologically-relevant analysis of the illumina infinium HumanMethylation450 BeadChip array. Epigenetics & chromatin, 6, 1. 292. Huang T., Hoffman B., Meschino W. and Okun N. (2010) Prediction of adverse pregnancy outcomes by combinations of first and second trimester biochemistry markers used in the routine prenatal screening of down syndrome. Prenat. Diagn., 30, 471-477. 293. Smith G.C., Stenhouse E.J., Crossley J.A., Aitken D.A., Cameron A.D. and Connor J.M. (2002) Early pregnancy levels of pregnancy-associated plasma protein a and the risk of 153  intrauterine growth restriction, premature birth, preeclampsia, and stillbirth. Journal of Clinical Endocrinology & Metabolism, 87, 1762-1767. 294. Ong C.Y., Liao A.W., Spencer K., Munim S. and Nicolaides K.H. (2000) First trimester maternal serum free β human chorionic gonadotrophin and pregnancy associated plasma protein A as predictors of pregnancy complications. BJOG: An International Journal of Obstetrics & Gynaecology, 107, 1265-1270. 295. Bersinger N.A., Smárason A.K., Muttukrishna S., Groome N.P. and Redman C.W. (2003) Women with preeclampsia have increased serum levels of pregnancy-associated plasma protein A (PAPP-A), inhibin A, activin A and soluble E-selectin. Hypertension in Pregnancy, 22, 45-55. 296. Sitras V., Paulssen R., Leirvik J., Vartun A. and Acharya G. (2009) Placental gene expression profile in intrauterine growth restriction due to placental insufficiency. Reprod. Sci., 16, 701-711. 297. Tsai S., Hardison N.E., James A.H., Motsinger-Reif A.A., Bischoff S.R., Thames B.H. and Piedrahita J.A. (2011) Transcriptional profiling of human placentas from pregnancies complicated by preeclampsia reveals disregulation of sialic acid acetylesterase and immune signalling pathways. Placenta, 32, 175-182. 298. Hogg K., Price E.M. and Robinson W.P. (2014) Improved reporting of DNA methylation data derived from studies of the human placenta. Epigenetics, 9, 333-337. 154  299. Nishizawa H., Ota S., Suzuki M., Kato T., Sekiya T., Kurahashi H. and Udagawa Y. (2011) Comparative gene expression profiling of placentas from patients with severe pre-eclampsia and unexplained fetal growth restriction. Reprod Biol Endocrinol, 9, 107. 300. Gurusinghe S., Wallace E.M. and Lim R. (2014) The relationship between activin A and anti-angiogenic factors in the development of pre-eclampsia. Pregnancy Hypertension: An International Journal of Women's Cardiovascular Health, 4, 3-6. 301. Yuen R.K., Peñaherrera M.S., von Dadelszen P., McFadden D.E. and Robinson W.P. (2010) DNA methylation profiling of human placentas reveals promoter hypomethylation of multiple genes in early-onset preeclampsia. European Journal of Human Genetics, 18, 1006-1012. 302. Tobi E.W., Lumey L.H., Talens R.P., Kremer D., Putter H., Stein A.D., Slagboom P.E. and Heijmans B.T. (2009) DNA methylation differences after exposure to prenatal famine are common and timing- and sex-specific. Hum. Mol. Genet., 18, 4046-4053. 303. El-Maarri O., Becker T., Junen J., Manzoor S.S., Diaz-Lacava A., Schwaab R., Wienker T. and Oldenburg J. (2007) Gender specific differences in levels of DNA methylation at selected loci from human total blood: A tendency toward higher methylation levels in males. Hum. Genet., 122, 505-514. 304. Engel S.M., Joubert B.R., Wu M.C., Olshan A.F., Haberg S.E., Ueland P.M., Nystad W., Nilsen R.M., Vollset S.E., Peddada S.D. et al. (2014) Neonatal genome-wide methylation 155  patterns in relation to birth weight in the norwegian mother and child cohort. Am. J. Epidemiol., 179, 834-842. 305. Auer J., Camoin L., Guillonneau F., Rigourd V., Chelbi S.T., Leduc M., Laparre J., Mignot T. and Vaiman D. (2010) Serum profile in preeclampsia and intra-uterine growth restriction revealed by iTRAQ technology. Journal of proteomics, 73, 1004-1017. 306. Akolekar R., Syngelaki A., Sarquis R., Zvanca M. and Nicolaides K.H. (2011) Prediction of early, intermediate and late pre!eclampsia from maternal factors, biophysical and biochemical markers at 11–13 weeks. Prenat. Diagn., 31, 66-74. 307. Overgaard M.T., Oxvig C., Christiansen M., Lawrence J.B., Conover C.A., Gleich G.J., Sottrup-Jensen L. and Haaning J. (1999) Messenger ribonucleic acid levels of pregnancy-associated plasma protein-A and the proform of eosinophil major basic protein: Expression in human reproductive and nonreproductive tissues. Biol. Reprod., 61, 1083-1089. 308. Lawrence J.B., Oxvig C., Overgaard M.T., Sottrup-Jensen L., Gleich G.J., Hays L.G., Yates J.R.,3rd and Conover C.A. (1999) The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A. Proc. Natl. Acad. Sci. U. S. A., 96, 3149-3153. 309. Mosesson M.W. and Amrani D.L. (1980) The structure and biologic activities of plasma fibronectin. Blood, 56, 145-158. 310. CLEMMENSEN I. (1981) Fibronectin and its role in connective tissue diseases. Eur. J. Clin. Invest., 11, 145-146. 156  311. Metzenbauer M., Hafner E., Hoefinger D., Schuchter K., Stangl G., Ogris E. and Philipp K. (2001) Three-dimensional ultrasound measurement of the placental volume in early pregnancy: Method and correlation with biochemical placenta parameters. Placenta, 22, 602-605. 312. Villar J., Carroli G., Wojdyla D., Abalos E., Giordano D., Ba'aqeel H., Farnot U., Bergsjø P., Bakketeig L. and Lumbiganon P. (2006) Preeclampsia, gestational hypertension and intrauterine growth restriction, related or independent conditions? Obstet. Gynecol., 194, 921-931. 313. Benton S.J., McCowan L.M., Heazell A.E., Grynspan D., Hutcheon J.A., Senger C., Burke O., Chan Y., Harding J.E. and Yockell-Lelièvre J. (2016) Placental growth factor as a marker of fetal growth restriction caused by placental dysfunction. Placenta, 42, 1-8. 314. American Academy of Pediatrics (2004) National high blood pressure education program working group on high blood pressure in children and adolescents. Pediatrics, 114, iv-iv. 315. Suzuki R. and Shimodaira H. (2006) Pvclust: An R package for assessing the uncertainty in hierarchical clustering. Bioinformatics, 22, 1540-1542. 316. Huang H., Liu Y., Marron J. and Huang M.H. (2015) Package ‘sigclust’, . 317. Gillis J., Mistry M. and Pavlidis P. (2010) Gene function analysis in complex data sets using ErmineJ. Nature protocols, 5, 1148-1159. 318. Yuen R.K., Chen B., Blair J.D., Robinson W.P. and Nelson D.M. (2013) Hypoxia alters the epigenetic profile in cultured human placental trophoblasts. Epigenetics, 8, 192-202. 157  319. Wallner W., Sengenberger R., Strick R., Strissel P.L., Meurer B., Beckmann M.W. and Schlembach D. (2007) Angiogenic growth factors in maternal and fetal serum in pregnancies complicated by intrauterine growth restriction. Clin. Sci. (Lond), 112, 51-57. 320. Benton S.J., Hu Y., Xie F., Kupfer K., Lee S., Magee L.A. and von Dadelszen P. (2012) Can placental growth factor in maternal circulation identify fetuses with placental intrauterine growth restriction? Am. J. Obstet. Gynecol., 206, 163. e1-163. e7. 321. Myatt L., Clifton R.G., Roberts J.M., Spong C.Y., Hauth J.C., Varner M.W., Thorp Jr J.M., Mercer B.M., Peaceman A.M. and Ramin S.M. (2012) First-trimester prediction of preeclampsia in nulliparous women at low risk. Obstet. Gynecol., 119, 1234-1242. 322. Jia R., Zhang X., Hu P., Liu X., Hua X., Wang X. and Ding H. (2012) Screening for differential methylation status in human placenta in preeclampsia using a CpG island plus promoter microarray. Int. J. Mol. Med., 30, 133. 323. Tobi E.W., Heijmans B.T., Kremer D., Putter H., Delemarre-van de Waal, Henriette A, Finken M.J., Wit J.M. and Slagboom P.E. (2011) DNA methylation of IGF2, GNASAS, INSIGF and LEP and being born small for gestational age. Epigenetics, 6, 171-176. 324. Roland C.S., Hu J., Ren C., Chen H., Li J., Varvoutis M.S., Leaphart L.W., Byck D.B., Zhu X. and Jiang S. (2016) Morphological changes of placental syncytium and their implications for the pathogenesis of preeclampsia. Cellular and molecular life sciences, 73, 365-376. 325. Houseman E.A., Molitor J. and Marsit C.J. (2014) Reference-free cell mixture adjustments in analysis of DNA methylation data. Bioinformatics, 30, 1431-1439. 158  326. Edgar R.D., Jones M.J., Robinson W.P. and Kobor M.S. (2017) An empirically driven data reduction method on the human 450K methylation array to remove tissue specific non-variable CpGs. Clinical Epigenetics, 9, 11. 327. Langfelder P. and Horvath S. (2008) WGCNA: An R package for weighted correlation network analysis. BMC Bioinformatics, 9, 559. 328. Peters T., Peters M.T., biocViews DifferentialMethylation G., GenomeAnnotation D., OneChannel T. and MultipleComparison Q. (2016) Package ‘DMRcate’.  329. Irizarry R.A., Aryee M., Bravo H.C., Hansen K.D., Jaffee H.A., Irizarry M.R.A., RUnit S., biocViews DNAMethylation E. and Infrastructure M. (2013) Package ‘bumphunter’.  330. Harley C.B., Vaziri H., Counter C.M. and Allsopp R.C. (1992) The telomere hypothesis of cellular aging. Exp. Gerontol., 27, 375-382. 331. De Meyer T., Rietzschel E.R., De Buyzere M.L., De Bacquer D., Van Criekinge W., De Backer G.G., Gillebert T.C., Van Oostveldt P., Bekaert S. and Asklepios investigators (2007) Paternal age at birth is an important determinant of offspring telomere length. Hum. Mol. Genet., 16, 3097-3102. 332. Blackburn E.H. (2005) Telomeres and telomerase: Their mechanisms of action and the effects of altering their functions. FEBS Lett., 579, 859-862. 333. Forsyth N.R., Wright W.E. and Shay J.W. (2002) Telomerase and differentiation in multicellular organisms: Turn it off, turn it on, and turn it off again. Differentiation, 69, 188-197. 159  334. Kyo S., Takakura M., Kanaya T., Zhuo W., Fujimoto K., Nishio Y., Orimo A. and Inoue M. (1999) Estrogen activates telomerase. Cancer Res., 59, 5917-5921. 335. Blackburn E. and Szostak J. (1984) The molecular structure of centromeres and telomeres. Annu. Rev. Biochem., 53, 163-194. 336. Prescott J., Kraft P., Chasman D.I., Savage S.A., Mirabello L., Berndt S.I., Weissfeld J.L., Han J., Hayes R.B. and Chanock S.J. (2011) Genome-wide association study of relative telomere length. PloS one, 6, e19635. 337. Codd V., Mangino M., van der Harst P., Braund P.S., Kaiser M., Beveridge A.J., Rafelt S., Moore J., Nelson C. and Soranzo N. (2010) Common variants near TERC are associated with mean telomere length. Nat. Genet., 42, 197-199. 338. Codd V., Nelson C.P., Albrecht E., Mangino M., Deelen J., Buxton J.L., Hottenga J.J., Fischer K., Esko T. and Surakka I. (2013) Identification of seven loci affecting mean telomere length and their association with disease. Nat. Genet., 45, 422-427. 339. Collopy L.C., Walne A.J., Cardoso S., de la Fuente J., Mohamed M., Toriello H., Tamary H., Ling A.J., Lloyd T. and Kassam R. (2015) Triallelic and epigenetic-like inheritance in human disorders of telomerase. Blood,126, 176-184. 340. Al Khaldi R., Mojiminiyi O., AlMulla F. and Abdella N. (2015) Associations of TERC single nucleotide polymorphisms with human leukocyte telomere length and the risk of type 2 diabetes mellitus. PloS one, 10, e0145721. 160  341. De Meyer T., Vandepitte K., Denil S., De Buyzere M.L., Rietzschel E.R. and Bekaert S. (2014) A non-genetic, epigenetic-like mechanism of telomere length inheritance? Eur. J. Hum. Genet., 22, 10-11. 342. Hansen R.S., Wijmenga C., Luo P., Stanek A.M., Canfield T.K., Weemaes C.M. and Gartler S.M. (1999) The DNMT3B DNA methyltransferase gene is mutated in the ICF immunodeficiency syndrome. Proc. Natl. Acad. Sci. U. S. A., 96, 14412-14417. 343. Yehezkel S., Segev Y., Viegas-Pequignot E., Skorecki K. and Selig S. (2008) Hypomethylation of subtelomeric regions in ICF syndrome is associated with abnormally short telomeres and enhanced transcription from telomeric regions. Hum. Mol. Genet., 17, 2776-2789. 344. Gonzalo S., Jaco I., Fraga M.F., Chen T., Li E., Esteller M. and Blasco M.A. (2006) DNA methyltransferases control telomere length and telomere recombination in mammalian cells. Nat. Cell Biol., 8, 416-424. 345. Cawthon R.M. (2002) Telomere measurement by quantitative PCR. Nucleic Acids Res., 30, e47. 346. Hanna C.W., Bretherick K.L., Gair J.L., Fluker M.R., Stephenson M.D. and Robinson W.P. (2009) Telomere length and reproductive aging. Hum. Reprod., 24, 1206-1211. 347. Ihaka R. and Gentleman R. (1996) R: A language for data analysis and graphics. Journal of computational and graphical statistics, 5, 299-314. 161  348. Cherif H., Tarry J.L., Ozanne S.E. and Hales C.N. (2003) Ageing and telomeres: A study into organ- and gender-specific telomere shortening. Nucleic Acids Res., 31, 1576-1583. 349. Benetos A., Dalgard C., Labat C., Kark J.D., Verhulst S., Christensen K., Kimura M., Horvath K., Kyvik K.O. and Aviv A. (2014) Sex difference in leukocyte telomere length is ablated in opposite-sex co-twins. Int. J. Epidemiol., 43, 1799-1805. 350. Biron-Shental T., Sukenik-Halevy R., Sharon Y., Goldberg-Bittman L., Kidron D., Fejgin M.D. and Amiel A. (2010) Short telomeres may play a role in placental dysfunction in preeclampsia and intrauterine growth restriction. Am. J. Obstet. Gynecol., 202, 381. e1-381. e7. 351. Meyerson M. (1998) Telomerase enzyme activation and human cell immortalization. Toxicol. Lett., 102, 41-45. 352. Guilleret I., Yan P., Grange F., Braunschweig R., Bosman F.T. and Benhattar J. (2002) Hypermethylation of the human telomerase catalytic subunit (hTERT) gene correlates with telomerase activity. International journal of cancer, 101, 335-341. 353. Watts J.M., Dumitriu B., Hilden P., Chen C., Rapaport F., Kishtagari A., Ahn J., Devlin S.M., Rampal R.K. and Levine R.L. (2014) Telomere length is associated with specific mutations and mutation classes in patients with acute myeloid leukemia. Blood, 124, 2280-2280. 354. Chen T., Ueda Y., Xie S. and Li E. (2002) A novel Dnmt3a isoform produced from an alternative promoter localizes to euchromatin and its expression correlates with active de novo methylation. J. Biol. Chem., 277, 38746-38754. 162  355. Oda M., Oxley D., Dean W. and Reik W. (2013) Regulation of lineage specific DNA hypomethylation in mouse trophectoderm. PloS one, 8, e68846. 356. Phillippe M. (2015) Cell-free fetal DNA, telomeres, and the spontaneous onset of parturition. Reprod. Sci., 22, 1186-1201. 357. Lappalainen T. and Greally J.M. (2017) Associating cellular epigenetic models with human phenotypes. Nature Reviews Genetics,18,441-451. 358. Davy P., Nagata M., Bullard P., Fogelson N. and Allsopp R. (2009) Fetal growth restriction is associated with accelerated telomere shortening and increased expression of cell senescence markers in the placenta. Placenta, 30, 539-542. 359. Januar V., Desoye G., Novakovic B., Cvitic S. and Saffery R. (2015) Epigenetic regulation of human placental function and pregnancy outcome: Considerations for causal inference. Obstet. Gynecol., 213, S182-S196. 360. Ernst J., Kheradpour P., Mikkelsen T.S., Shoresh N., Ward L.D., Epstein C.B., Zhang X., Wang L., Issner R. and Coyne M. (2011) Mapping and analysis of chromatin state dynamics in nine human cell types. Nature, 473, 43-49. 361. Geeleher P., Hartnett L., Egan L.J., Golden A., Raja Ali R.A. and Seoighe C. (2013) Gene-set analysis is severely biased when applied to genome-wide methylation data. Bioinformatics, 29, 1851-1857. 163  362. Suderman M., Pappas J.J., Borghol N., Buxton J.L., McArdle W.L., Ring S.M., Hertzman C., Power C., Szyf M. and Pembrey M. (2015) Lymphoblastoid cell lines reveal associations of adult DNA methylation with childhood and current adversity that are distinct from whole blood associations. Int. J. Epidemiol., 44, 1331-1340. 363. Joubert B.R., Felix J.F., Yousefi P., Bakulski K.M., Just A.C., Breton C., Reese S.E., Markunas C.A., Richmond R.C. and Xu C. (2016) DNA methylation in newborns and maternal smoking in pregnancy: Genome-wide consortium meta-analysis. The American Journal of Human Genetics, 98, 680-696. 364. Chuang Y., Quach A., Absher D., Assimes T., Horvath S. and Ritz B. (2017) Coffee consumption is associated with DNA methylation levels of human blood. European Journal of Human Genetics, 25, 608-616. 365. Plant D., Webster A., Nair N., Oliver J., Smith S.L., Eyre S., Hyrich K.L., Wilson A.G., Morgan A.W. and Isaacs J.D. (2016) Differential methylation as a biomarker of response to etanercept in patients with rheumatoid arthritis. Arthritis & Rheumatology, 68, 1353-1360. 366. Smith A.K., Kilaru V., Kocak M., Almli L.M., Mercer K.B., Ressler K.J., Tylavsky F.A. and Conneely K.N. (2014) Methylation quantitative trait loci (meQTLs) are consistently detected across ancestry, developmental stage, and tissue type. BMC Genomics, 15, 145. 367. Lam L.L., Emberly E., Fraser H.B., Neumann S.M., Chen E., Miller G.E. and Kobor M.S. (2012) Factors underlying variable DNA methylation in a human community cohort. Proc. Natl. Acad. Sci. U. S. A., 109 Suppl 2, 17253-17260. 164  368. Saunders C.L., Abel G.A., El Turabi A., Ahmed F. and Lyratzopoulos G. (2013) Accuracy of routinely recorded ethnic group information compared with self-reported ethnicity: Evidence from the english cancer patient experience survey. BMJ Open, 3, 10.1136/bmjopen-2013-002882. 369. Halder I., Shriver M., Thomas M., Fernandez J.R. and Frudakis T. (2008) A panel of ancestry informative markers for estimating individual biogeographical ancestry and admixture from four continents: Utility and applications. Hum. Mutat., 29, 648-658. 370. Rahmani E., Shenhav L., Schweiger R., Yousefi P., Huen K., Eskenazi B., Eng C., Huntsman S., Hu D. and Galanter J. (2017) Genome-wide methylation data mirror ancestry information. Epigenetics & chromatin, 10, 1. 371. Wang J., Han X. and Sun Y. (2017) DNA methylation signatures in circulating cell-free DNA as biomarkers for the early detection of cancer. Science China Life Sciences, , 1-7. 372. Warton K. and Samimi G. (2015) Methylation of cell-free circulating DNA in the diagnosis of cancer. Frontiers in Molecular Biosciences, 2, 13. 373. Keravnou A., Ioannides M., Tsangaras K., Loizides C., Hadjidaniel M.D., Papageorgiou E.A., Kyriakou S., Antoniou P., Mina P. and Achilleos A. (2016) Whole-genome fetal and maternal DNA methylation analysis using MeDIP-NGS for the identification of differentially methylated regions. Genetics research, 98. 165  374. Kazemi M., Salehi M. and Kheirollahi M. (2017) MeDIP real-time qPCR has the potential for noninvasive prenatal screening of fetal trisomy 21. International Journal of Molecular and Cellular Medicine (IJMCM), 6, 13-21.  166  Appendices  Appendix A  Supplementary tables and figures for Chapter 1  Supplementary Table A.1. Summary of preeclampsia and intrauterine growth restriction DNA methylation studies with both genome-wide and targeted approaches. Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%GENOMEEWIDE%ANALYSES%2017% Kim,%Jeong/Hyun,%et%al.%"Genome/wide%DNA%methylation%profiles%of%maternal%peripheral%blood%and%placentas:%potential%risk%factors%for%preeclampsia%and%validation%of%GRK5."%Genes%&%Genomics:%1/10.%PE%Control%PE/%BP>140/90%mmHg%and%proteinuria%>300mg/day%or%>2+%dipstick)%after%20%weeks%gestation.%PE%(N=6)%Control%(N=6)%p<0.01,%Δβ>0.17%Not%clear% Matched%gestational%age%365%differentially%methylated%sites%were%identified%between%PE%and%controls%Illumina%450K%Array%No% NA%2016% Roifman,%Maian,%et%al.%"Genome/wide%placental%DNA%methylation%analysis%of%severely%growth/discordant%monochorionic%twins%reveals%novel%epigenetic%targets%for%intrauterine%Discordant%monozygotic%twins%for%IUGR%IUGR/%Fetal%birthweight%<%10th%percentile%Monozygotic%monchorionic%twin%pairs%disconcordant%for%p<0.05,%∆β>0.1,%and%more%than%3%CpG%sites%in%the%region%Not%clear% Fetal%sex%Maternal%age%Gestational%age%Bump%Hunting:%Increased%DNAm%in%DECR1,%ZNF300,%LEPR,%HSPA1A/L,GST01,GNE.%Decreased%DNAm%in%DNAJ4%and%CCL28.%Array/wide:%296%differentially%Illumina%450K%Array%No% NA%167  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%growth%restriction."%Clinical%Epigenetics%8.1%(2016):%70.%IUGR%(N=8)%methylated%CpG%sites%Sites%associated%with%fetal%development,%intracellular%signalling,%and%cell%adhesion%%2016% Yeung,%Kristen%Rose,%et%al.%"DNA%methylation%profiles%in%preeclampsia%and%healthy%control%placentas."%American%Journal%of%PhysiologyLHeart%and%Circulatory%Physiology%(2016):%ajpheart/00958.%PE%Control%PE/%BP>140/90%mmHg%after%20%weeks%gestation,%proteinuria%(>2+%dipstick%or%300mg/24h),%or%renal%insufficiency,%liver%disease,%neurologic%problems,%haematological%disturbances.%All%were%late/onset%PE.%Control/%Uncomplicated%term%pregnancies%PE%(N=8)%Control%(N=16)%Adjusted%p<0.05%Yes% Gestational%age%At%an%adjusted%p%value%<0.05,%no%site%on%the%array%was%found%to%be%significant%between%PE%and%Control.%303%differentially%methylated%regions%were%identified%using%DMRcate%between%PE%and%controls%after%adjusting%for%gestational%age.%92%of%these%site%(34.7%)%remained%significant%in%the%secondary%cohort.%Illumina%450K%Array%Yes% Sites%were%validated%on%an%independent%cohort%from%GEO%(Blair%et%al%2013).%%2016% He,%Zhiming,%et%al.%"The%promoter%methylomes%of%monochorionic%twin%placentas%reveal%intrauterine%Monchorionic%twins%discordant%for%IUGR%IUGR/%fetal%birth%weight%<10th%percentile%for%gestational%age%Discovery%(N=7%pairs)%Validation%p<0.05,%∆β>0.2%Not%to%identify%initial%differentially%methylated%sites.%%None% Genome%wide%promoter%methylation%was%significantly%lower%in%the%IUGR%placenta%Bisulfite%genomic%sequencing%enriched%for%No% NA%168  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%growth%restriction/specific%variations%in%the%methylation%patterns."%Scientific%reports%6%(2016).%(N=10%pairs)%Differentially%methylated%sites%were%merged%into%candidate%DMRs%based%on%defined%criteria.%Candidate%DMRs%were%considered%differentially%methylated%is%they%met%FDR<0.05%compared%to%the%non/IUGR%twin.%4,605%sites%were%differentially%methylated%between%IUGR%and%AGA%twin.%promoter%regions%(1.86%million%sites)%2015% Hillman,%Sara%L.,%et%al.%"Novel%DNA%methylation%profiles%associated%with%key%gene%regulation%and%transcription%pathways%in%blood%and%placenta%of%growth/restricted%neonates."%Epigenetics%10.1%(2015):%50/61.%IUGR%Adequate%for%gestational%age%(AGA)%IUGR/%fetal%birth%weight%<10th%percentile%for%gestational%age,%adjusted%for%maternal%weight,%height,%parity%and%ethnicity.%%AGA/%appropriately%growth%fetuses%between%>10th/<95th%percentile,%born%>255%day%gestation.%IUGR%(N=23)%AGA%(N=22)%FDR<0.05% Yes% Gestational%age%Not%accounting%for%gestational%age,%454%differentially%methylated%sites%were%identified%between%IUGR%and%AGA%placentas.%Illumina%450K%Array%No% NA%2015% Zhu,%Lisha,%et%al.%"Genome/wide%mapping%of%5mC%and%5hmC%identified%LOPE%Control%PE/%BP>%140/90%mmHg,%proteinuria%>+2%or%300mg/24h,%all%samples%were%>37%weeks%LOPE%(N=4)%Control%(N=4)%>2%fold,%p<0.05%No% None% 714%sites%were%found%to%be%differentially%methylated%in%MeDIP% No% NA%169  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%differentially%modified%genomic%regions%in%late/onset%severe%preeclampsia:%a%pilot%study."%PloS%one%10.7%(2015):%e0134119.%gestation%Control/%Healthy%uncomplicated%control%LOPE%compared%to%controls.%2014% Blair,%J.%D.,%et%al.%"Overlapping%DNA%methylation%profile%between%placentas%with%trisomy%16%and%early/onset%preeclampsia."%Placenta%35.3%(2014):%216/222.%1st%trimester%control%1st%trimester%trisomy%16%3rd%trimester%control%3rd%trimester%CPM16%1st%trimester%control/%chromosomally%normal,%termination%or%spontaneous%abortion%<12%weeks%gestation%1st%trimester%trisomy%16/%terminations%or%spontaneous%abortion%with%confirmed%trisomy%16%<12%weeks%gestation%3rd%trimester%control/%included%both%term%(>37%weeks)%and%preterm%(<37%weeks)%placentas.%Preterm%births%were%due%to%a%variety%of%etiologies%3rd%trimester%1st%trimester%control%(N=5)%1st%trimester%trisomy%16%(N=5)%3rd%trimester%control%(N=10)%3rd%trimester%CPM16%(N=10)%Different%FDR%thresholds%(0.01,0.05,0.1,%1)%and%different%∆β%thresholds%(0%,5%,10%,15%)%were%investigated.%%Yes% Gestational%age%Widespread%DNAm%alterations%were%reported%between%CPM16%and%3rd%trimester%controls.%A%significant%overlap%of%altered%DNAm%in%previously%reported%in%EOPE%was%found%in%the%sites%differentially%methylated%in%CPM16%placentas.%Illumina%450K%Array%Yes% DNAm%alteration%were%followed%up%with%pyrosequencing%on%an%independent%set%of%8%CPM16%and%37%3rd%trimester%control%170  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%CPM16/%placentas%with%confirmed%CPM16%>24%weeks%gestation.%2014% Anton,%Lauren,%et%al.%"Differential%methylation%of%genes%associated%with%cell%adhesion%in%preeclamptic%placentas."%PLoS%One%9.6%(2014):%e100148.%Term%PE%Preterm%PE%Term%Control%Term%PE/%diagnosis%≥37%weeks%gestation%Preterm%PE/%diagnosis%≤37%weeks%gestation%Term%Control/%Healthy%placenta%≥37%weeks%gestation%Term%PE%(N=19)%Preterm%PE%(N=12)%Term%Control%(N=14)%Preterm%PE%vs%Term%Control/%FDR<0.01,%∆β>0.05%Other%thresholds%were%also%looked%at%Yes% None% At%FDR<0.01,%∆β>0.05,%421%CpG%sites%were%differentially%methylated%between%Preterm%PE%and%term%controls.%None%were%differentially%methylated%between%term%PE%and%term%controls%%Illumina%450K%Array%Select%sites%were%confirmed%with%pyrosequencing/%not%clear%if%these%are%in%the%same%samples%or%independent%cohort.%NA%171  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2014% Ching,%Travers,%et%al.%"Genome/wide%hypermethylation%coupled%with%promoter%hypomethylation%in%the%chorioamniotic%membranes%of%early%onset%pre/eclampsia."%Molecular%human%reproduction%20.9%(2014):%885/904.%Early/onset%PE%(EOPE)%Term%Controls%EOPE/%PE%diagnosis%≤34%weeks%gestation%Term%controls/%Healthy%placenta%≥37%weeks%gestation%EOPE%(N=30)%Term%controls%(N=17)%Bonferroni%adjusted%p/value<0.05%Top%hits%used%∆β>0.2%Yes% None% Found%that%of%sites%that%met%Bonferroni%correction,%there%was%an%enrichment%for%global%hypermethylation%and%enrichment%of%hypomethylation%at%promoter%regions%of%genes%Illumina%450K%Array%No% NA%2014% Anderson,%Cindy%M.,%et%al.%"DNA%methylation%as%a%biomarker%for%preeclampsia."%Biological%research%for%nursing%16.4%(2014):%409/420.%PE%Control%PE/%BP>%140/90%mmHg,%proteinuria%>+1%or%300mg/24h%Control/%Healthy%uncomplicated%control%PE%(N=6)%Control%(N=6)%p/value%<0.05,%Δβ>0.2%No%% None% Found%206%sites%that%were%differentially%methylated%in%maternal%blood.%Of%which%64%%of%those%that%were%hypermethylated,%and%24%%that%were%hypomethylated%were%also%altered%in%placental%tissue.%Illumina%450K%Array%No% NA%172  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2014% Chu,%Tianjiao,%et%al.%"Comprehensive%analysis%of%preeclampsia/associated%DNA%methylation%in%the%placenta."%PloS%one%9.9%(2014):%e107318.%PE%Control%PE/%BP>%140/90%mmHg,%proteinuria%>+1%(catherizied%sample),%2+%(dipstick)%or%300mg/24h%Control/%Healthy%uncomplicated%control%PE%(N=24)%Control%(N=24)%FDR<0.1,%Δβ>0.1%%Yes% None% 28%differentially%methylated%CpG%sites%were%identified%between%PE%and%control%when%X%chromosome%probes%were%removed%from%the%analysis.%11%differentially%methylated%sites%were%identified%when%analysis%was%only%completed%on%female%placentas.%Illumina%27K%array%Partially% Results%were%confirmed%by%Sequenom%epityper%Epityper%%and%bisulfite%sequencing.%Results%were%compared%to%previously%reported%alterations%in%the%literature%(Yuen%et%al%2010%and%Blair%et%al%2013).%With%little%overlap%2014% Nomura,%Yoko,%et%al.%"Global%methylation%in%the%placenta%and%umbilical%cord%blood%from%pregnancies%with%maternal%gestational%diabetes,%preeclampsia,%and%obesity."%Reproductive%sciences%21.1%(2014):%131/137.%Gestational%Diabetes%PE%Obesity%Control%Gestational%diabetes/%GCT%>100mg/dL%or%GTT%>95mg/dL%fasting%or%>180,155,and%140mg/dL%at%1,2,3%hours,%respectively.%%PE/%BP>140/90%mmHg,%proteinuria%>300mg/24h%or%+2%dipstick%after%20%weeks%gestation%Obesity/%pre/pregnancy%maternal%BMI%>30%Ns%not%specified%p%value<0.05%Not%clear% Not%specified%Global%DNAm%was%significantly%lower%in%gestational%diabetes%and%PE%placentas%than%controls.%Global%DNAm%was%lower%in%obesity%placentas%than%Controls%LUMA%Assay%No% No%173  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%Control/%wasn't%defined%2013% Blair,%John%D.,%et%al.%"Widespread%DNA%hypomethylation%at%gene%enhancer%regions%in%placentas%associated%with%early/onset%preeclampsia."%Molecular%human%reproduction%(2013):%gat044.%Early/onset%PE%(EOPE)%Control%EOPE/%PE%diagnosis%prior%to%34%weeks%gestation%Control/%gestationally%age%matched%control.%Preterm%births%due%to%a%variety%of%aetiologies.%EOPE%(N=20)%Controls%(N=20)%FDR<0.01,%Δβ>0.125%Yes% Birth%weight%Gestational%age%Head%circumference%Body%length%at%birth%Found%significant%hypomethylation%in%enhancers%and%low%density%CpG%regions%in%EOPE%placentas.%Illumina%450K%Array%No% NA%2013% Yan,%Y.%H.,%et%al.%"Screening%for%preeclampsia%pathogenesis%related%genes."%Eur%Rev%Med%Pharmacol%PE%Control%PE/%definition%not%specified,%all%samples%3rd%trimester%%Control/%uncomplicated%PE%(N=30)%Control%(N=30)%p%value<0.05%No% None% %Candidate%genes%were%chosen%from%sites%that%were%differentially%methylated%(specifics%of%DNA%methylation%array%not%specified%No% NA%174  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%Sci%17.22%(2013):%3083/3094.%pregnancy%in%3rd%trimester%analysis%not%given).%2011% Banister,%Carolyn%E.,%et%al.%"Infant%growth%restriction%is%associated%with%distinct%patterns%of%DNA%methylation%in%human%placentas."%Epigenetics%6.7%(2011):%920/927.%SGA%and%IUGR%Term%Controls%SGA/IUGR/%Fetal%birthweight%<%10th%percentile%for%its%gestational%age%Term%Control/%Healthy%placenta%≥36%weeks%gestation%SGA%&%IUGR%(N=89)%Term%Controls%(N=117)%Not%specified%No% Fetal%sex%Maternal%age%Mode%of%delivery%22%loci%identified%in%the%training%data%set%predicted%76%%of%SGA%and%IUGR%placentas%in%the%testing%dataset%Illumina%27K%Array%Yes% Cohort%was%divided%into%a%testing%and%training%dataset%2010% Liu,%Lan,%et%al.%"Distinct%DNA%methylomes%of%human%placentas%between%pre/eclampsia%and%gestational%diabetes%mellitus."%Cellular%Physiology%and%Biochemistry%34.6%(2014):%1877/1889.%Control%Gestation%Diabetes%PE%Control/%uncomplicated%term%pregnancies%Gestational%Diabetes/%75g%oral%glucose%tolerance%test%resulted%in%glucose%level%of%≥5.1mmol/L%glucose%and/or%at%2%hours%plasma%glucose%level%≥8.1mmol/L%glucose.%PE/%BP%≥140/90%mmHg%and%proteinuria%≥2+%dipstick%or%≥300mg/24h%Control%(N=30)%Gestational%Diabetes%(N=%28)%PE%(N=27)%p%value<0.05%Yes% None% DNAm%profiles%in%both%PE%and%gestational%diabetes%were%altered.%64.4%%of%the%DNAm%alterations%were%shared%in%both%PE%and%gestational%diabetes.%These%sites%were%enriched%for%pathways%such%as%cell%adhesion%and%cell%differentiation%and%metabolism.%%Wasn't%clear:%Either%NimbleGen%microarray%or%Illumina%27K%array%No% NA%175  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2010% Yuen,%Ryan%KC,%et%al.%"DNA%methylation%profiling%of%human%placentas%reveals%promoter%hypomethylation%of%multiple%genes%in%early/onset%preeclampsia."%European%Journal%of%Human%Genetics%18.9%(2010):%1006/1012.%Control%Early/onset%PE%(EOPE)%Late/onset%PE%(LOPE)%IUGR%Controls/%uncomplicated%pregnancy.%Divided%into%early%controls%(preterm%births%due%to%varying%aetiologies),%late%controls%and%controls%(mean%gestational%age%37).%EOPE/%diagnosis%of%PE%<%34%weeks%gestation.%LOPE/%diagnosis%of%PE%>34%weeks%gestation.%IUGR/%fetal%birth%weight%<%3rd%percentile,%or%<%10th%percentile%with%other%indications%of%poor%growth.%Early%control%(N=4)%Late%control%(N=5)%Control%(N=5)%EOPE%(N=4)%LOPE%(N=4)%IUGR%(N=4)%FDR<0.1,%Δβ>0.1%%Yes% Matched%gestational%age%Gene%specific%hypomethylation%was%observed%in%EOPE,%but%not%LOPE%or%IUGR.%%Golden%gate%array%Yes% 4%loci%found%to%be%hypomethylated%in%EOPE%were%validated%by%pyrosequencing%in%an%independent%cohort.%%%%%%%%%%%%%%%%%%%%%%%%%176  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%TARGETED%ANALYSES%2017% Rahat,%Beenish,%et%al.%"The%role%of%aberrant%methylation%of%trophoblastic%stem%cell%origin%in%the%pathogenesis%and%diagnosis%of%placental%disorders."%Prenatal%Diagnosis%37.2%(2017):%133/143.%PE%1st%trimester%control%2nd%trimester%control%3rd%trimester%control%Molar%pregnancy%PE/%BP>140/90%mmHg%and%proteinuria%>300mg/24h%All%Controls/%singleton,%uncomplicated,%pregnancy%Molar/%Diagnosed%by%ultrasound%and%confirmed%with%histropathology%PE%(N=30)%1st%trimester%control%(N=30)%2nd%trimester%control%(N=30)%3rd%trimester%control%(N=30)%Molar%(N=15)%p%value<0.05%No% Matched%gestational%age%EFGR%promoter%region%was%<2.5%%methylated%in%each%of%the%study%groups,%with%no%significant%differences%between%them.%VEGF%and%cLjun%promoter%regions%were%differentially%methylated.%PE%was%increased%at%VEGF%and%cLjunpromoters.%Molar%pregnancies%were%hypomethylated%at%VEGF%and%cLjun%promoters.%%MS/HRM% No% NA%177  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2017% Majchrzak/Celińska,%Aleksandra,%et%al.%"HSD11B2,%RUNX3,%and%LINE/1%Methylation%in%Placental%DNA%of%Hypertensive%Disorders%of%Pregnancy%Patients."%Reproductive%Sciences%(2017):%1933719117692043.%Gestational%Hypertension%PE%Chronic%hypertension%Normotensive%Control%Gestational%hypertension/%hypertension%after%20%weeks%gestation,%and%normalization%of%BP%pospartum.%PE/%BP>140/90%mmHg%and%proteinuria%(>0.3g/24h)%after%20%weeks%gestation.%Chronic%hypertension/%prepregnancy%hypertension%or%hypertension%<20%weeks%gestation.%Control/%normotensive%women,%no%proteinuria%and%no%IUGR.%Gestational%hypertension%(N=13)%PE%(N=11)%Chronic%hypertension%(N=10)%Normotensive%control%(N=25)%p%value<0.05%No% Gestational%age%Maternal%age%Fetal%sex%All%models%were%corrected%for%gestational%age%and%fetal%sex.%By%linear%regression,%no%differences%in%DNA%methylation%in%HSD11B2,%RUNX3,%LINEL1%were%observed%between%PE,%gestational%hypertension,%chronic%hypertension%and%control.%%1%CpG%in%HSD11B2%was%significantly%increased%in%PE%compared%to%control%with%a%student's%t/test,%however%the%effect%went%away%once%gestational%age%and%fetal%sex%were%corrected%for.%Methylation%specific%PCR%and%pyrosequencing%No% NA%178  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2017% Xiao,%Xirong,%et%al.%"Hypomethylation%of%tissue%factor%pathway%inhibitor%2%in%human%placenta%of%preeclampsia."%Thrombosis%Research%152%(2017):%7/13.%PE%Control%PE/%onset%of%hypertension%in%previously%normotensive%women%and%proteinuria%(>0.3g/24h)%Control/%uncomplicated%healthy%pregnancies,%undergoing%elective%caesarean.%%PE%(N=19)%Control%(N=10)%p%value<0.05%No% None% DNA%methylation%levels%were%altered%along%the%TFP1L2%gene%in%PE%placentas%compared%to%controls.%MassArray%EPITYPER%No% NA%2016% Ye,%Weiping,%et%al.%"Preeclampsia%is%Associated%with%Decreased%Methylation%of%the%GNA12%Promoter."%Annals%of%human%genetics%80.1%(2016):%7/10.%PE%Control%PE/%BP>140/90%mmHg%and%proteinuria%(>300mg/24h)%after%20%weeks%gestation.%%Control/%Uncomplicated%healthy%pregnancy%PE%(N=50)%Control%(N=50)%p%value<0.05%No% None% DNA%methylation%was%significantly%lower%in%3%of%the%8%CpG%sites%analyzed%within%the%GNA12%promoter%in%PE%compared%to%controls.%When%dividing%PE%group%into%mild%vs%severe%or%early/onset%and%late/onset,%no%significant%differences%in%DNA%methylation%were%observed.%MassArray%EPITYPER%No% NA%179  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2016% Lin,%Liang,%et%al.%"Significant%hypomethylation%of%TNFAIP8%and%increased%expression%in%the%placenta%and%peripheral%blood%cells%from%early/onset%preeclamptic%patients."%INTERNATIONAL%JOURNAL%OF%CLINICAL%AND%EXPERIMENTAL%MEDICINE%9.6%(2016):%10384/+.%Early/onset%PE%(EOPE)%Late/onset%PE%(LOPE)%Term%Control%EOPE/%PE%diagnosis%<%34%weeks%gestation%LOPE/%PE%diagnosis%>34%weeks%gestation%Term%control/%healthy%pregnancy%delivered%at%term.%% p%value<0.05%No% None% Two%sites%in%the%promoter%region%of%TNFAIP8%were%assessed.%DNA%methylation%at%site%1%was%significantly%different%between%LOPE%and%control%and%LOPE%and%EOPE.%No%other%significant%differences%were%found.%Pyrosequencing%No% No%2015% Anderson,%Cindy%M.,%et%al.%"First%trimester%vitamin%D%status%and%placental%epigenomics%in%preeclampsia%among%Northern%Plains%primiparas."%Life%sciences%129%(2015):%10/15.%PE%Control%This%was%a%prospective%study,%PE%was%defined%at%BP>%140/90%mmHg%and%proteinuria%(>300mg%in%24h)%PE%(N=3)*%Control%(N=3)*%*Samples%which%DNAm%was%measured%%None%specified%No% None% DNAm%was%increased%at%CYP27B1,%VDR,%and%RXR%(Involved%in%vitamin%D%metabolism)%promoter%regions.%Human%DNA%Methylation%2.1%M%Microarray%(NimbleGen)%No% NA%2015% Tang,%Yao,%et%al.%"Hypermethylation%of%the%HLA/G%promoter%is%associated%with%preeclampsia."%MolPE%Control%PE%/%BP<%140/90%mmHg%after%20%weeks%gestation%and%proteinuria%(>300mg/24h%or%>1+%dipstick).%PE%(N=19)%Control%(N=20)%p%value<0.05%No% None% HLA/G%promoter%shows%differential%methylation%at%some%CpG%sites%in%PE%placentas%Pyrosequencing%No% NA%180  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%ecular%human%reproduction%21.9%(2015):%736/744.%Control/%uncomplicated%pregnancies%undergoing%elective%cesarian%compared%to%controls.%2015% Iglesias/Platas,%I.,%et%al.%"Distinct%promoter%methylation%and%isoform/specific%expression%of%RASFF1A%in%placental%biopsies%from%complicated%pregnancies."%Placenta%36.4%(2015):%397/402.%IUGR%Adequate%for%gestational%age%(AGA)%IUGR%was%classified%by%most%severe%ultrasound%abnormality.%PE%was%included%in%this%cohort%AGA/Normally%grown%fetuses%IUGR%(N=47)%AGA%(N=67)%p%value<0.05%Yes% Gestational%age%A%CpG%site%in%the%RASSF1A%promoter%was%hypermethylated%in%IUGR%placentas%compared%to%controls.%The%DNA%methylation%change%was%confirmed%in%a%subset%of%IUGR%and%AGA%that%were%run%on%the%array%using%pyrosequencing.%Illumina%Goldengate%array%No% NA%2015% Xiao,%Xirong,%et%al.%"Fetal%growth%restriction%and%methylation%of%growth/related%genes%in%the%placenta."%(2015).%IUGR%Control%IUGR/%fetal%growth%<10th%percentile%for%gestational%age,%based%on%ultrasound%measures.%Control/%Appropriately%grown%fetuses%IUGR%(N=80)%Control%(N=101)%p%value<0.05%No% Gestational%age%DNA%methylation%within%HSD11B2%was%significantly%decreased%in%IUGR%placentas%compared%to%controls,%after%correction%for%gestational%age,%prenatal%vitamin%use,%and%maternal%smoking%status.%%Pyrosequencing%No% NA%181  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2014% Lu,%Linshan,%et%al.%"Methylation%pattern%of%H19%exon%1%is%closely%related%to%preeclampsia%and%trophoblast%abnormalities."%International%journal%of%molecular%medicine%34.3%(2014):%765/771.%First%trimester%controls%Term%Control%Severe%PE%First%trimester%controls%collected%between%6/9%weeks%Term%Control/%Health%placenta%in%the%third%trimester%Severe%PE/%Hypertension%(160/110%mmHg)%and%proteinuria%after%20%weeks%gestation.%Sample%collected%in%third%trimester%First%trimester%(N=6)%Term%Placenta%(N=16)%Severe%PE%(N=15)%p%value<0.05%No% None% DNAm%at%7%CpG%sites%(averaged)%within%the%first%exon%of%H19%were%hypomethylated%in%first%trimester%placenta%compared%to%term.%The%average%methylation%across%the%region%was%not%different%between%term%and%severe%PE%placentas.%Looking%at%the%7%CpG%sites%individually,%only%DNAm%at%CpG%site%1%was%significantly%different%between%term%placenta%and%severe%PE.%No%other%comparisons%were%significant.%%Pyrosequencing%No% NA%182  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2014% Liu,%Q.,%et%al.%"Promoter%hypomethylation%and%increased%maspin%expression%in%preeclamptic%placentas%in%a%Chinese%population."%Placenta%35.11%(2014):%876/882.%PE%Control%PE/definition%not%specified%Control/%uncomplicated%pregnancies%PE%(N=12)%Control%(N=12)%p%value<0.05%No% None% DNA%methylation%in%the%MSP%promoter%was%significantly%decreased%in%PE%placentas%compared%to%controls.%No%differences%were%observed%between%male%and%female%placentas.%Methylation%specific%PCR%and%pyrosequencing%Maybe% 24%placental%samples%were%assessed%for%DNA%methylation%at%the%MSP%promoter%region.%MSP%was%again%hypomethylated%in%the%PE%group%compared%to%control.%It%is%unclear%if%this%is%an%independent%cohort%or%a%subset%of%the%original%cohort.%2014% Zhao,%Yan,%et%al.%"Site/specific%methylation%of%placental%HSD11B2%gene%promoter%is%related%to%intrauterine%growth%restriction."%European%Journal%of%Human%Genetics%22.6%(2014):%734/740.%IUGR%Control%IUGR/%diagnosed%IUGR%if%they%had%3%of%the%following%ultrasound%findings:%1.%abdominal%circumference<3rd%percentile%for%gestational%age,%2.%abnormal%umbilical%artery%Doppler,%3.%oligohydramnios,%or%4.%abnormal%fetal%growth%velocity<1.5%IUGR%(N=22)%(Some%also%had%PE)%Control%(N=22)%p%value<0.05%No% Gestational%age%Fetal%sex%Maternal%age%Maternal%BMI%Smoking%Pregnancy%weigh%gain%Promoter%DNA%methylation%of%HSDB112%were%increased%in%IUGR%placentas%compared%to%controls.%After%correction%for%clinical%variables,%only%2%CpG%sites%in%the%promoter%were%associated%with%birth%weight.%%%Pyrosequencing%No% NA%183  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%standard%deviations%over%14%days.%%Control/%healthy%pregnancies,%adequate%fetal%growth%2013% Sundrani,%Deepali%P.,%et%al.%"Differential%placental%methylation%and%expression%of%VEGF,%FLT/1%and%KDR%genes%in%human%term%and%preterm%preeclampsia."%Clinical%epigenetics%5.1%(2013):%6.%Term%PE%Preterm%PE%Term%Control%Term%PE/%diagnosis%≥37%weeks%gestation%Preterm%PE/%diagnosis%≤37%weeks%gestation%Term%Control/%Healthy%placenta%≥37%weeks%gestation%Term%PE%(N=48)%Preterm%PE%(N=45)%Term%Control%(N=46)%p%value<0.05%No% None% Mean%promoter%DNAm%at%VEGF%promoter%hypomethylated%in%preterm%PE%compared%to%term%controls.%%No%significant%difference%in%either%FLT1%or%KDR%promoter%regions.%Hypomethylation%at%single%sites%within%FLT1%in%both%term%and%preterm%PE.%Hypermethylation%in%single%site%in%KDR%promoter%in%both%term%and%preterm%PE.%Sequenom%Epityper%No% NA%184  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2013% Xiang,%Yuqian,%et%al.%"Up/regulated%expression%and%aberrant%DNA%methylation%of%LEP%and%SH3PXD2A%in%pre/eclampsia."%PloS%one%8.3%(2013):%e59753.%PE%Uncomplicated%Pregnancy%PE/%BP%140/90%mmHG%&%proteinuria%>0.3mg/24h%Uncomplicated%pregnancy/%normal%pregnancy%PE%(N=16)%Uncomplicated%pregnancy%(N=16)%p%value<0.05%No% Gestational%age%DNAm%at%transcriptional%binding%sites%Sp1,LP1,%and%CEBPα,%as%well%as%DNAm%at%transcriptional%start%site%of%LEP,%and%several%gene%body%CpG%sites%in%SH3PXD2A%were%differentially%methylated.%%Sequenom%Epityper%No% NA%2013% Xiang,%Yuqian,%et%al.%"Promoter%hypomethylation%of%TIMP3%is%associated%with%pre/eclampsia%in%a%Chinese%population."%Molecular%human%reproduction%19.3%(2013):%153/159.%PE%Control%PE/%BP>140/90%mmHg%and%proteinuria%(>0.3g/24h)%after%20%weeks%gestation%Control/%healthy%pregnancies%undergoing%caesarean%PE%(N=41)%Control%(N=22)%p%value<0.05%No% None% The%TIMP3%promoter%was%hypomethylated%in%PE%compared%to%controls%in%the%Han%Chinese%population.%As%it%was%in%the%Canadian%population%MassArray%EPITYPER%Yes% This%study%was%itself%a%validation%of%the%TIMP3%promoter%methylation%reported%in%Yuen%et%al%(2010),%using%an%independent%cohort.%185  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2013% Hogg,%Kirsten,%et%al.%"Early%onset%pre/eclampsia%is%associated%with%altered%DNA%methylation%of%cortisol/signalling%and%steroidogenic%genes%in%the%placenta."%PloS%one%8.5%(2013):%e62969.%Early/onset%PE%(EOPE)%Late/onset%PE%(LOPE)%normotensive%IUGR%(nIUGR)%Control%EOPE/%PE%diagnosis%≤34%weeks%gestation%LOPE/%PE%diagnosis>34%weeks%gestation%nIUGR/%Fetal%birth%weight<%3rd%percentile%or%<10th%percentile%with%other%indications%of%poor%growth%Control/%Uncomplicated%pregnancies%with%no%known%placental%pathology.%Some%of%these%samples%include%preterm%birth%due%to%a%number%of%aetiologies%not%such%as:%premature%rupture%of%membranes,%incompetent%cervix,%and%chorioamnionitis.%EOPE%(N=19)%LOPE%(N=18)%nIUGR%(N=13)%Control%(N=111)%p%value<0.01,%Δβ>0.05%Yes% Birth%weight,%gestational%age,%maternal%age,%fetal%sex,%presence/absence%of%labour%all%assessed,%confounder%included%if%statistically%significant%DNAm%was%increased%within%NR3C1,%CRHBP,%and%CRH%%in%EOPE%compared%to%control.%No%differential%DNAm%was%observed%in%HSD11B2%gene%promoter.%DNAm%was%decreased%CYP11A1,%HSD3B1,%TEAD3,%and%CYP19%in%EOPE%and%LOPE,%but%not%nIUGR.%Pyrosequencing%Partially%(in%EOPE)%Initial%differences%in%DNAm%were%observed%between%EOPE%and%Controls%using%the%Illumina%450K%array.%This%subset%of%samples%were%expanded%on%and%LOPE%and%nIUGR%samples%added%in%and%DNAm%assessed%via%pyrosequencing.%2013% Janzen,%Carla,%et%al.%"Placental%glucose%transporter%3%(GLUT3)%is%up/regulated%in%human%pregnancies%IUGR%PE%Control%IUGR/%fetal%birth%weight%<10th%percentile%and%other%indication%of%poor%growth%PE/%definition%not%IUGR%(N=10)%PE%(N=5)%Control%(N=10)%p%value<0.05%No% None% No%differences%in%DNA%methylation%were%found%in%GLUT3%in%IUGR%or%PE%placentas%Pyrosequencing%No% NA%186  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%complicated%by%late/onset%intrauterine%growth%restriction."%Placenta%34.11%(2013):%1072/1078.%specified%Control%uncomplicated%pregnancy%compared%to%controls%2011% Ferreira,%Jose%Carlos,%et%al.%"WNT2%promoter%methylation%in%human%placenta%is%associated%with%low%birthweight%percentile%in%the%neonate."%Epigenetics%6.4%(2011):%440/449.%SGA%Control%SGA/%fetal%birth%weight%<10th%persentile%Control/%fetal%weight%appropriate%for%gestational%age%SGA%(N=8)%Control%(N=8)%p%value<0.05%No% Matched%gestational%age%WNT2%promoter%DNA%methylation%was%showed%1.9x%fold%change%across%the%promoter%in%SGA%vs%controls%(p%value%not%given),%using%the%Agilent%array.%%Change%in%DNA%methylation%at%the%WNT2%promoter%was%confirmed%in%the%same%samples%using%pyrosequencing.%MeDIP%and%244K%Agilent%Microarray%No% NA%2011% Koukoura,%O.,%et%al.%"Hypomethylation%along%with%increased%H19%expression%in%placentas%from%pregnancies%complicated%with%fetal%growth%restriction."%Control%Fetal%growth%restriction%(FGR)%Control/%uncomplicated%term%placenta%FGR/%Fetal%birth%weight%<%5th%percentile%in%addition%to%other%indications%of%poor%growth.%Control%(N=17)%FGF%(N=31)%p%value<0.05%No% Birth%weight%Gestational%age%DNAm%at%two%regions%within%H19%were%measured.%DNAm%between%6156bp/6245bp%upstream%of%the%TSS%were%not%significantly%different,%although%trended%MethylLight%Assay%No% NA%187  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%Placenta%32.1%(2011):%51/57.%to%be%hypomethylated%in%FGF.%Between%7705bp%and%7809bp,%close%to%the%promoter,%there%was%significant%hypomethylation%in%FGF%compared%to%controls.%No%correlation%between%DNAm%and%gene%expression%was%observed.%2011% Koukoura,%Ourania,%et%al.%"Loss%of%imprinting%and%aberrant%methylation%of%IGF2%in%placentas%from%pregnancies%complicated%with%fetal%growth%restriction."%International%journal%of%molecular%medicine%28.4%(2011):%481.%IUGR%Control%IUGR/%fetal%birth%weight%<5th%percentile%and%other%indications%of%poor%fetal%growth%Control/%Term,%uncomplicated,%appropriate%for%gestational%age%growth%between%>10th/<90th%percentiles.%IUGR%(N=31)%Control%(N=17)%p%value<0.05%No% Birth%weight%Gestational%age%No%difference%in%DNA%methylation%observed%at%IGF2between%IUGR%and%control%placentas%MethylLight%Assay%No% NA%188  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%2011% Gao,%Wen/long,%et%al.%"Detection%of%global%DNA%methylation%and%paternally%imprinted%H19%gene%methylation%in%preeclamptic%placentas."%Hypertension%Research%34.5%(2011):%655/661.%Early/onset%PE%(EOPE)%Late/onset%PE%(LOPE)%Controls%EOPE/%diagnosis%of%PE%<34%weeks%LOPE/%diagnosis%of%PE%>34%weeks%Controls/%not%specified%but%appear%to%be%term%EOPE%(N=10)%LOPE%(N=14)%Control(%N=24)%p<0.01,%Δβ>0.17%No% None% Global%DNAm,%measured%by%both%immunohistrochemistry%and%LINE1%and%ALU%pyrosequencing%revealed%higher%DNAm%levels%in%EOPE%than%controls.%LINE1%and%Alu%were%also%significantly%higher%in%LOPE.%H19%DNAm,%measured%by%MS/HRM,%was%also%reported%higher%in%EOPE%than%controls%immunohistorchemitry%LINE1%and%ALU%pyrosequencing%MS/HRM%No% NA%2011% Lambertini,%Luca,%et%al.%"Differential%methylation%of%imprinted%genes%in%growth/restricted%placentas."%Reproductive%sciences%18.11%(2011):%1111/1117.%IUGR%Adequete%for%gestational%age%(AGA)%IUGR/%fetal%birth%weight%<5th%percentile%and%associated%with%either%absent%end/diastolic%slow%or%reverse%end%diastolic%slow%of%the%umbliical%artery%AGA/%pregnancies%deliverying>37%weeks%gestation,%>10th%percentile%in%fetal%birth%weight%and%no%signs%of%IUGR%(N=7)%AGA%(N=10)%p<0.01% No% None% Imprinted%genes%in%the%placenta%had%a%trend%toward%hypermethylation%in%IUGR%placentas.%MeDIP% No% NA%189  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%growth%restriction%by%ultrasound%2010% Bourque,%D.%K.,%et%al.%"Decreased%placental%methylation%at%the%H19/IGF2%imprinting%control%region%is%associated%with%normotensive%intrauterine%growth%restriction%but%not%preeclampsia."%Placenta%31.3%(2010):%197/202.%Control%IUGR%PE%PE%+%IUGR%Control/%Uncomplicated%term%placenta%IUGR/%fetal%birth%weight%<%3rd%percentile%or%<10th%percentile%with%other%indicators%of%poor%clinical%growth.%PE/%BP≥%140/90%mmHg%after%20%weeks%gestation%with%either%proteinuria%(300mg/24h),%or%other%clinical%features%defined%in%SOGC.%PE%+%IUGR/%PE%with%fetal%birth%weight%<%3rd%percentile%or%<10th%percentile%with%other%Control%(N=22)%IUGR%(N=13)%PE%(N=17)%PE%+%IUGR%(N=21)%p%value<0.05%Yes% None% DNAm%at%ICR1,%associated%with%H19%and%IGF2%gene%expression%was%reduced%in%nIUGR%placentas,%but%not%PE.%DNAm%at%ICR2,%associated%with%KvDMR1%and%CDKN1C%gene%expression,%was%not%significantly%altered%in%either%PE%or%nIUGR%placentas.%Illumina%Goldengate,%pyrosequencing,%MS/SNuPE%assay%Yes% Initial%DNAm%was%measured%on%a%small%set%of%samples%using%the%golden%gate%array.%DNAm%was%validated%with%a%larger%cohort%using%the%Ms/SNuPE%(ICR1)%and%pyrosequencing%(ICR2)%190  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%indicators%of%poor%clinical%growth.%2010% Wang,%Zhuoqun,%et%al.%"Expressional%and%epigenetic%alterations%of%placental%matrix%metalloproteinase%9%in%preeclampsia."%Gynecological%Endocrinology%26.2%(2010):%96/102.%PE%Control%PE/%BP%140/90%mm%HG%&%proteinuria%>0.3mg/24h%Uncomplicated%pregnancy/%normal%pregnancy%Control/Health%uncomplicated%third%trimester%placenta%PE%(N=20)%Control%(N=18)%p%value<0.05%No% None% A%CpG%site%at%/712bp%from%the%transcriptional%start%site%showed%altered%DNAm%in%the%PE%group%compared%to%the%control.%Magnitude%of%change%was%not%reported.%Methylation%sensitive%restriction%enzyme%digest,%followed%by%PCR%No% NA%2008% Guo,%Lin,%et%al.%"Altered%gene%expression%and%methylation%of%the%human%chromosome%11%imprinted%region%in%small%for%gestational%age%(SGA)%SGA%Controls%IUGR%SGA/%>26%weeks%of%gestation%with%fetal%growth%<10th%percentile%Control/%fetal%birth%weight>10th%persentile,%without%any%complications%IUGR/%SGA%placentas%with%SGA%(N=24,%2%were%twins)%Control(N=20,%8%were%twins)%IUGR%(N=15,%p%value<0.05%No% None% KvDMR%and%H19%were%not%differentially%methylated%in%IUGR%compared%to%AGA%controls.%One%SGA%sample%showed%loss%of%imprinting.%Methylation%sensitive%enzyme%digestion%with%southern%blotting%No% NA%191  Year% Citation% Pathology% Pathology%Definition% N=% Thresholds% Multiple%test%corrections%Confounders%% Findings% Platform% Validated% How%was%it%validated?%placentae."%Developmental%biology%320.1%(2008):%79/91.%maternal%vascular%underperfusion,%or%distal%villous%hypoplasia%these%were%also%in%the%SGA%group)%2007% Chelbi,%Sonia%T.,%et%al.%"Expressional%and%epigenetic%alterations%of%placental%serine%protease%inhibitors."%Hypertension%49.1%(2007):%76/83.%PE%IUGR%PE+IUGR%Control%PE/%BP>140/90%mmHg%and%proteinuria%>0.3g%per%24h.%IUGR/%fetal%birth%weight%<10th%percentile%Control/%Uncomplicated%pregnancy%undergoing%caesarean%section%PE%(N=9)%PE+IUGR%(N=7)%IUGR%(N=8)%Control(N=7)%p<0.05% No% None% Promoter%region%of%SERPINA3%is%hypomethylated%in%PE%compared%to%control.%Sequencing%and%pyrosequencing%Yes% DNA%methylation%in%the%promoter%of%SERPINA3%was%validated%in%a%larger%secondary%cohort%consisting%of%47%PE,16%IUGR,%and%18%control%placentas.%192  Appendix B  Supplementary tables and figures for Chapter 2 Supplementary Table B.1. List of probes filtered from the discovery and validation cohorts. 193   Appendix C  Supplementary tables and figures for Chapter 3 Supplementary Table C.1 All samples used in Chapter 3 and what measures were completed on each sample. SAMPLE'ID' GROUP' 450K'Gene'expression' Pyrosequencing' FN1'ELISA'PAPPA'(MSS)'INHA'(MSS)'PL135' LOPE+IUGR* X* ** ** ** ** **PL142' LOPE+IUGR* X* ** ** ** ** **PM31' LOPE+IUGR* X* ** X* X* ** **PM32' LOPE+IUGR* X* ** X* X* ** **PM38' LOPE+IUGR* X* ** X* X* ** **PM40' LOPE+IUGR* X* ** X* X* ** **PM52' LOPE+IUGR* X* ** X* X* ** **PM66' LOPE+IUGR* X* ** X* X* ** **PM115' LOPE** X* ** X* X* ** **PM119' LOPE** X* ** X* X* ** **PM44' LOPE** X* ** X* X* ** **PM46' LOPE** X* ** X* X* ** **PM53' LOPE** X* ** X* X* ** **PM54' LOPE** X* ** X* X* ** **PM55' LOPE** X* ** X* X* ** **PM58' LOPE** X* ** X* X* ** **PM71' LOPE** X* ** X* X* ** **PM98' LOPE** X* ** X* X* ** **PM269' LOPE* X* ** ** ** ** **PL145' IUGR* X* ** ** ** ** **PL72' IUGR* X* ** ** ** ** **PL86' IUGR* X* ** ** ** ** **PM121' IUGR* ** ** X* X* ** **PM123' IUGR* X* ** X* X* ** **PM128' IUGR* ** ** X* X* ** **PM130' IUGR* X* ** X* X* ** **PM139' IUGR* X* ** X* X* ** **PM29' IUGR* X* ** ** ** ** **194  PM30' IUGR* X* ** X* X* ** **PM35' IUGR* X* ** ** ** ** **PM4' IUGR* X* ** X* X* ** **PM41' IUGR* X* ** ** ** ** **PM42' IUGR* ** ** X* X* ** **PM47' IUGR* X* ** X* X* ** **PM72' IUGR* X* ** ** ** ** **PL130' EOPE* X* ** ** ** ** **PL131' EOPE* X* ** ** ** ** **PM116' EOPE* X* X* X* X* ** **PM12' EOPE* X* X* ** ** ** **PM129' EOPE* X* X* X* X* ** **PM138' EOPE* X* ** ** ** ** **PM15' EOPE* X* ** X* X* ** **PM21' EOPE* X* X* X* X* ** **PM36' EOPE* X* ** X* X* ** **PM39' EOPE* X* X* X* X* ** **PM43' EOPE* X* X* X* X* ** **PM49' EOPE* X* ** ** ** ** **PM50' EOPE* ** ** X* X* ** **PM51' EOPE* X* X* ** ** ** **PM6' EOPE* X* ** X* X* ** **PM60' EOPE* ** ** X* X* ** **PM62' EOPE* ** ** X* X* ** **PM64' EOPE* X* ** X* X* ** **PM67' EOPE* X* ** ** ** ** **PM80' EOPE* X* ** X* X* ** **PM86' EOPE* X* X* ** ** ** **PM97' EOPE* X* ** ** ** ** **PM99' EOPE* X* ** ** ** ** **PL102' CONTROL* X* ** ** ** ** **PL104' CONTROL* X* ** ** ** ** **PL11' CONTROL* X* X* ** ** ** **PL112' CONTROL* X* ** ** ** ** **PL113' CONTROL* X* X* ** ** ** **PL21' CONTROL* X* ** ** ** ** **PL25' CONTROL* X* ** ** ** ** **PL26' CONTROL* X* ** ** ** ** **PL32' CONTROL* X* ** ** ** ** **195  PL33' CONTROL* X* X* ** ** ** **PL38' CONTROL* X* ** ** ** ** **PL43' CONTROL* X* ** ** ** ** **PL56' CONTROL* X* X* ** ** ** **PL58' CONTROL* X* ** ** ** ** **PL59' CONTROL* X* ** ** ** ** **PL64' CONTROL* X* X* ** ** ** **PL65' CONTROL* X* X* ** ** ** **PL76' CONTROL* X* X* ** ** ** **PL96' CONTROL* X* X* ** ** ** **PM'266' CONTROL* ** ** X* ** ** **PM'268' CONTROL* ** ** X* ** ** **PM100' CONTROL* ** ** X* X* ** **PM101' CONTROL* ** ** X* X* ** **PM103' CONTROL* ** ** X* X* ** **PM106' CONTROL* ** ** X* X* ** **PM110' CONTROL* ** ** X* X* ** **PM111' CONTROL* ** ** X* X* ** **PM112' CONTROL* X* ** X* X* ** **PM113' CONTROL* ** ** X* X* ** **PM114' CONTROL* X* ** X* X* ** **PM117' CONTROL* ** ** X* X* ** **PM118' CONTROL* ** ** X* X* ** **PM120' CONTROL* X* ** X* X* ** **PM122' CONTROL* X* ** X* X* ** **PM124' CONTROL* ** ** X* X* ** **PM127' CONTROL* ** ** X* X* ** **PM134' CONTROL* ** ** X* X* ** **PM136' CONTROL* X* ** X* X* ** **PM142' CONTROL* X* ** ** ** ** **PM153' CONTROL* X* ** ** ** ** **PM158' CONTROL* X* ** ** ** ** **PM161' CONTROL* X* ** ** ** ** **PM163' CONTROL* ** ** X* X* ** **PM167' CONTROL* X* ** ** ** ** **PM17' CONTROL* X* ** ** ** ** **PM176' CONTROL* ** ** X* ** X* **PM179' CONTROL* ** ** X* X* ** **PM190' CONTROL* ** ** X* X* ** X*PM191' CONTROL* ** ** X* X* ** X*196  PM193' CONTROL* ** ** X* X* ** X*PM198' CONTROL* ** ** X* X* X* **PM20' CONTROL* X* ** ** ** ** **PM200' CONTROL* ** ** X* X* ** X*PM201' CONTROL* ** ** X* X* ** X*PM202' CONTROL* ** ** X* X* ** **PM205' CONTROL* X* ** ** ** ** **PM209' CONTROL* ** ** X* ** ** X*PM212' CONTROL* ** ** X* X* X* X*PM214' CONTROL* ** ** X* X* X* X*PM214K2' CONTROL* ** ** X* X* ** X*PM215' CONTROL* ** ** X* X* X* X*PM216' CONTROL* ** ** X* X* X* X*PM217' CONTROL* ** ** X* X* X* X*PM218' CONTROL* ** ** X* X* X* **PM219' CONTROL* ** ** X* X* X* X*PM220' CONTROL* ** ** X* X* X* X*PM221' CONTROL* ** ** X* X* ** X*PM222' CONTROL* ** ** X* X* X* X*PM223' CONTROL* ** ** X* X* X* X*PM224' CONTROL* ** ** X* X* X* X*PM227' CONTROL* ** ** X* ** X* X*PM228' CONTROL* ** ** X* X* X* X*PM229' CONTROL* ** ** X* X* X* X*PM232' CONTROL* ** ** X* X* X* X*PM234' CONTROL* ** ** X* X* X* X*PM236' CONTROL* ** ** X* ** X* X*PM237' CONTROL* ** ** X* ** X* X*PM239' CONTROL* ** ** X* X* X* **PM243' CONTROL* ** ** X* X* X* X*PM248' CONTROL* ** ** X* X* ** X*PM249' CONTROL* X* ** X* ** X* X*PM252' CONTROL* ** ** X* X* X* X*PM254' CONTROL* ** ** X* X* ** X*PM258' CONTROL* ** ** X* X* X* X*PM259' CONTROL* ** ** X* X* X* X*PM260' CONTROL* ** ** X* X* X* X*PM261' CONTROL* ** ** X* X* X* **PM262' CONTROL* ** ** X* X* X* X*197  PM263' CONTROL* X* ** ** ** ** **PM264' CONTROL* ** ** X* X* X* **PM266' CONTROL* ** ** X* X* X* X*PM267' CONTROL* ** ** X* X* X* **PM268' CONTROL* ** ** X* X* X* X*PM57' CONTROL* ** ** X* X* ** **PM59' CONTROL* ** ** X* X* ** **PM65' CONTROL* ** ** X* X* ** **PM70' CONTROL* ** ** X* X* ** **PM74' CONTROL* X* ** X* X* ** **PM77' CONTROL* X* ** X* X* ** **PM78' CONTROL* ** ** X* X* ** **PM79' CONTROL* ** ** X* X* ** **PM81' CONTROL* ** ** X* X* ** **PM83' CONTROL* ** ** X* X* ** **PM84' CONTROL* ** ** X* X* ** **PM85' CONTROL* ** ** X* X* ** **PM87' CONTROL* X* ** X* X* ** **PM88' CONTROL* ** ** X* X* ** **PM89' CONTROL* ** ** X* X* ** **PM90' CONTROL* ** ** X* X* ** **PM91' CONTROL* ** ** X* X* ** **PM93' CONTROL* ** ** X* X* ** **PM94' CONTROL* ** ** X* X* ** **PM96' CONTROL* ** ** X* X* ** ** 198   Supplementary'Table'C.2.''Samples used to assess the relationship between DNA methylation and gene expression in the term placenta in Chapter 3.   Control EOPE N= 8 8  Mean GA at delivery (weeks ±SD) 31.5 (±3.9) 32.2 (±3.9)  Mean BW (grams ±SD) 1885.0 (±677.2) 1420.0 (±394.8)  Mean  MA (years ±SD) 30.2 (±5.9) 33.4 (±5.9)  Sex (Female/N, %) 2/8, 25% 2/8, 25% † Previously published in Blair et al. 2013.199    Supplementary Table C.3. Primer sequences for bisulfite pyrosequencing in Chapter 3. Specific locations are based on UCSC hg/18 assembly.  Gene Primer Sequence Genomic Target PCR conditions DNA loaded into Pyro INHBA Forward GGAGTGGGAGATAAGGTTTATGGTTTA Chr2: 220436857-220436999 95(15')/95(30"),*55(30"),*72(30")*x40/72(10')** 5uL  Reverse ACACCCACCCTCTTCTACCCT     Sequencing GGTTTATGGTTATAGATATT      CpG 1 Chr2: 220436894   PAPPA Forward GGGATTTATTTAATTGGGAAGGTG Chr9: 118915879-118915998 95(15')/95(30"),*55(30"),*72(30")*x40/72(10')**8uL  Reverse CCTCCCTCCTTCTTCTCAAA    200   Sequencing TTAATTGGGAAGGTGG      CpG 1 118915907     CpG 2 118915929   FN1 Forward TGGGGGAAAGAGGTTATGATAT Chr 2: 216402317-216402556 95(15')/95(30"),*55(30"),*72(30")*x40/72(10') 8uL  Reverse ACAAAAATTACCAAATCCTAAACATTTACT     Sequencing TGAGTTGTATTTTTTGGAAGT      CpG 1  216402384     201   Supplementary Figure C.1. A Spearman’s correlation was performed between CpG 1 and CpG 2 within the PAPPA assay. DNA methylation was averaged over the two CpGs.202  Appendix D  Supplementary figures and tables for Chapter 4  Supplementary Figure D.1. Sample placement on the 96 well plates (Plate) and the microarray chips for the discovery cohort.203   Supplementary Figure D.2. Sample placement on the 96 well plates (Plate) and the microarray chips for the validation cohort.204   Supplementary Figure D.3. Sample-sample correlations for both the discovery and validation cohort in the control samples (both term and preterm). P-value from Student’s t-test indicates that the discovery cohort is more heterogeneous (less correlated) than the validation cohort.205   Supplementary Figure D.4. Sample-sample correlations for both the discovery and validation cohort in the EOPE samples. P-value from Student’s t-test indicates that the discovery cohort is more heterogeneous (less correlated) than the validation cohort.206  Supplementary Table D.1. CpG site and gene information on the 599 validated hits.  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg13056990) chr1) AJAP1) AJAP1) 20995) 0.10) 3.95E204) 0.06) 3.81E201)cg15492834) chr1) C1orf226) C1orf226) 2462) 0.10) 3.10E207) 0.11) 2.51E203)cg25212701) chr1) C1orf226) C1orf226) 145) 0.11) 7.67E206) 0.10) 2.71E202)cg21582311) chr1) C1orf96) C1orf96) 3380) 0.13) 1.80E207) 0.10) 1.99E203)cg10620457) chr1) C8B) C8B) 2422) 0.10) 4.79E206) 0.05) 1.84E202)cg24141382) chr1) CTPS) CTPS) 2874) 0.10) 1.21E205) 0.09) 4.01E202)cg08885854) chr1) DNM3) 7SK) 255636) 0.12) 3.83E206) 0.02) 1.61E202)cg14072140) chr1) DPYD) DPYD) 354776) 0.12) 5.06E204) 0.00) 4.28E201)cg07179075) chr1) DPYD) PTBLP) 426808) 0.10) 2.78E204) 0.07) 3.21E201)cg17850498) chr1) ECE1) AX747766) 39113) 0.19) 1.91E215) 0.02) 2.30E207)cg05797770) chr1) EPHA2) EPHA2) 11915) 0.11) 6.90E206) 0.02) 2.58E202)cg14178899) chr1) ERRFI1) ERRFI1) 2847) 0.13) 2.85E208) 0.04) 5.89E204)cg22671342) chr1)FAIM3;IL24) IL24) 6552) 0.10) 2.15E208) 0.01) 4.58E204)cg21837826) chr1) FAM163A) FAM163A) 4434) 0.11) 3.03E205) 0.02) 8.20E202)cg03730249) chr1) FOXJ3) GUCA2A) 261618) 0.12) 1.96E206) 0.03) 9.70E203)cg23677911) chr1) GALNT2) GALNT2) 53439) 0.17) 4.58E212) 0.13) 4.06E206)cg26020161) chr1) GSTM3) GSTM3) 2763) 0.11) 1.52E206) 0.08) 7.88E203)cg04103490) chr1) HIST3H3) HIST3H3) 21092) 0.13) 6.24E208) 0.11) 9.81E204)cg09664054) chr1) HIVEP3) HIVEP3) 2128108) 0.14) 1.62E208) 0.01) 3.53E204)cg16175792) chr1) HSD3B1) HSD3B1) 279) 0.15) 7.88E211) 0.05) 1.61E205)cg08872493) chr1) HTR1D) HTR1D) 2195) 0.12) 2.47E207) 0.03) 2.30E203)cg11811391) chr1) HTR1D) HTR1D) 1139) 0.16) 2.27E208) 0.11) 4.74E204)207  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg20803547) chr1) IL12RB2) IL12RB2) 394) 20.11) 7.40E203) 20.03) 9.64E201)cg07177437) chr1) ITLN1) ITLN1) 4583) 0.11) 1.10E207) 0.10) 1.43E203)cg01567509) chr1) KIAA1026) KAZN) 2146401) 0.10) 1.84E205) 0.03) 5.65E202)cg05046721) chr1) KIAA1026) KAZN) 242369) 0.12) 6.05E207) 0.12) 4.25E203)cg08538034) chr1) KIAA1026) KAZN) 3733) 0.12) 8.09E209) 0.08) 2.42E204)cg22507756) chr1) KIAA1026) KAZN) 2639) 0.12) 2.55E206) 0.07) 1.17E202)cg24589834) chr1) KIAA1026) KAZN) 157880) 0.11) 1.53E206) 0.14) 8.05E203)cg27538859) chr1) KIAA1026) KAZN) 2143924) 0.12) 6.83E208) 0.06) 1.09E203)cg07344172) chr1) KIF21B) KIF21B) 25666) 0.11) 1.10E205) 0.04) 3.62E202)cg07913153) chr1) LGALS8) LGALS8) 2752) 0.10) 3.34E209) 0.15) 1.46E204)cg27182012) chr1) LMNA) LMNA) 2803) 0.17) 5.76E208) 0.09) 9.22E204)cg01334432) chr1)LOC642587;MIR205) MIR205) 2894) 0.10) 9.88E210) 0.03) 6.54E205)cg12701302) chr1) MAN1C1) MAN1C1) 23710) 0.14) 4.88E206) 0.04) 1.90E202)cg01869896) chr1) MFSD2A) MFSD2A) 2484) 0.10) 1.16E207) 0.10) 1.47E203)cg17966362) chr1) MSH4) MSH4) 2756) 0.14) 7.78E211) 0.02) 1.56E205)cg10994126) chr1) PAPPA2) PAPPA2) 192) 0.16) 1.97E212) 0.06) 3.34E206)cg18236464) chr1) PAPPA2) PAPPA2) 96477) 0.20) 4.99E210) 0.03) 4.11E205)cg25103772) chr1) PAPPA2) PAPPA2) 81399) 0.15) 1.45E211) 0.08) 8.02E206)cg16056164) chr1) PCNXL2) PCNXL2) 217308) 0.11) 1.33E205) 0.11) 4.32E202)cg00964446) chr1) PPFIA4) PPFIA4) 22274) 0.10) 2.63E206) 0.12) 1.19E202)cg03511671) chr1) PPFIA4) PPFIA4) 2790) 0.13) 7.49E208) 0.13) 1.19E203)cg00762397) chr1) PRELP) PRELP) 9128) 0.11) 1.41E208) 0.02) 3.47E204)cg08106205) chr1) PRKCZ) PRKCZ) 214131) 0.12) 7.38E209) 0.07) 2.34E204)cg06826449) chr1) PTPN14) PTPN14) 84398) 0.11) 5.71E206) 0.07) 2.32E202)208  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg26185836) chr1) SLC2A1) FLJ32224) 232029) 0.10) 1.05E211) 0.04) 6.14E206)cg25469314) chr1) SLC5A9) SLC5A9) 2162) 0.12) 3.34E205) 0.01) 8.67E202)cg16926302) chr1)ST6GALNAC5) ST6GALNAC5) 77721) 0.10) 2.24E203) 0.01) 7.16E201)cg20991819) chr1) TGFB2) U3) 97472) 0.11) 5.89E209) 0.12) 2.14E204)cg16565154) chr1)TNFAIP8L2)TNFAIP8L22SCNM1) 242) 0.11) 1.63E206) 0.07) 8.64E203)cg16903016) chr1) )) 7SK) 66873) 0.13) 2.28E207) 0.08) 2.20E203)cg26548682) chr1) )) AK123450) 22942) 0.11) 2.20E207) 0.08) 2.20E203)cg27307465) chr1) )) BC035370) 24643) 0.15) 1.06E208) 0.04) 2.83E204)cg10092779) chr1) )) BC048114) 21492) 0.12) 1.77E208) 0.06) 3.99E204)cg13897222) chr1) )) C1orf140) 59852) 0.14) 1.23E205) 0.18) 4.06E202)cg05371908) chr1) )) CACHD1) 236888) 0.11) 7.88E207) 0.10) 4.63E203)cg12792952) chr1) )) CACHD1) 279088) 0.14) 7.78E208) 0.08) 1.19E203)cg21283167) chr1) )) FOXD2) 84557) 0.10) 3.60E207) 0.01) 2.86E203)cg17108958) chr1) )) HLX) 76803) 0.11) 7.05E206) 0.13) 2.61E202)cg13173305) chr1) )) IFFO2) 239474) 0.11) 1.51E209) 0.03) 9.22E205)cg09332996) chr1) )) KAZN) 23363) 0.13) 1.41E211) 0.01) 7.64E206)cg13107839) chr1) )) KAZN) 277501) 0.11) 1.18E206) 0.03) 6.23E203)cg27589988) chr1) )) KCNK2) 230993) 0.10) 4.15E205) 0.06) 1.08E201)cg07817055) chr1) )) KIAA1383) 2107774) 0.10) 2.14E203) 0.06) 7.08E201)cg15109744) chr1) )) KIAA1614) 20308) 0.14) 8.51E211) 0.04) 1.67E205)cg26365254) chr1) )) KIRREL) 210944) 0.10) 2.76E205) 0.09) 7.35E202)cg25099516) chr1) )) LOC388630) 173749) 0.10) 1.17E205) 0.07) 3.82E202)cg07950244) chr1) )) PHLDA3) 210432) 0.19) 1.39E212) 0.04) 2.58E206)209  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg13196578) chr1) )) PLEKHM2) 22659) 0.10) 8.83E207) 0.22) 5.83E203)cg05819416) chr1) )) PROX12AS1) 2890) 0.10) 3.90E205) 0.08) 9.58E202)cg07915117) chr1) )) PTP4A2) 217143) 0.11) 9.46E209) 0.09) 2.61E204)cg27056501) chr1) )) QSOX1) 23267) 0.12) 1.03E205) 0.09) 3.36E202)cg24340655) chr1) )) RGS4) 282770) 0.12) 2.09E206) 0.06) 1.05E202)cg03650068) chr1) )) RHD) 22489) 0.10) 3.70E208) 0.07) 6.68E204)cg26922451) chr1) )) SLC30A2) 23553) 0.13) 2.23E208) 0.08) 4.67E204)cg11283860) chr1) )) SLC45A1) 2111037) 0.15) 1.05E210) 0.15) 1.74E205)cg02494582) chr10) ADAM12) ADAM12) 15292) 0.14) 8.58E210) 0.14) 6.15E205)cg23963517) chr10)BICC1;LOC728640) LOC728640) 2902) 0.14) 3.05E204) 0.07) 3.46E201)cg26159090) chr10)BICC1;LOC728640) LOC728640) 2522) 0.11) 1.33E205) 0.08) 4.37E202)cg01140247) chr10) C10orf11) Metazoa_SRP) 217702) 0.12) 2.20E208) 0.09) 4.61E204)cg04507071) chr10) C10orf90) C10orf90) 41530) 0.10) 1.40E207) 0.03) 1.62E203)cg04804772) chr10) CALHM3) CALHM3) 124) 0.10) 6.50E210) 0.03) 4.44E205)cg13553455) chr10) COL17A1) COL17A1) 2364) 0.14) 4.05E207) 0.10) 3.32E203)cg07125981) chr10) CUL2) CUL2) 32762) 0.11) 1.93E204) 0.03) 2.63E201)cg12750431) chr10) CXCL12) DKFZp547J0510) 22817) 0.11) 2.36E206) 0.08) 1.10E202)cg18474072) chr10) DIP2C) DL489896) 22980) 0.15) 7.36E208) 0.04) 1.16E203)cg00779924) chr10) FLJ45983) BC036297) 771) 0.11) 3.08E205) 0.05) 8.25E202)cg01522692) chr10) GATA3) GATA3) 20349) 0.10) 5.91E210) 0.05) 4.25E205)cg07989490) chr10) GATA3) GATA3) 20360) 0.10) 2.43E209) 0.05) 1.13E204)cg15258080) chr10) HK1) HK1) 12602) 0.15) 1.66E210) 0.10) 2.06E205)cg27321175) chr10) ITGA8) ITGA8) 72761) 0.11) 1.52E205) 0.07) 4.98E202)210  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg02448743) chr10) KAZALD1) KAZALD1) 1744) 0.11) 2.10E210) 0.10) 2.32E205)cg11693581) chr10) KIAA1598) KIAA1598) 2509) 0.11) 5.41E206) 0.07) 2.21E202)cg03932039) chr10) LHPP) LHPP) 2402) 0.11) 2.34E208) 0.08) 4.93E204)cg03865281) chr10) LIPJ) LIPJ) 18153) 0.11) 1.79E206) 0.09) 9.16E203)cg19192065) chr10)LOC100188947) LOC100188947) 36126) 0.10) 3.83E207) 0.10) 3.10E203)cg23697796) chr10)LOC100188947) LOC100188947) 36243) 0.13) 3.24E207) 0.08) 2.73E203)cg26673784) chr10)LOC283050) AX747983) 223937) 0.10) 5.70E206) 0.01) 2.28E202)cg20621765) chr10) MYOF) MYOF) 43700) 0.11) 1.52E206) 0.10) 7.92E203)cg19024161) chr10) PBLD) PBLD) 11830) 0.11) 6.92E208) 0.11) 1.10E203)cg14500854) chr10) PRKG1) PRKG1) 252335) 0.11) 3.79E205) 0.02) 9.23E202)cg25306480) chr10) PWWP2B) PWWP2B) 15816) 0.12) 7.54E209) 0.03) 2.35E204)cg25961733) chr10) PWWP2B) PWWP2B) 17538) 0.11) 6.07E209) 0.01) 2.15E204)cg19140548) chr10)SH3PXD2A) LOC100505839) 45870) 0.19) 2.84E211) 0.03) 1.02E205)cg14920044) chr10) SLC16A12) SLC16A12) 2998) 0.10) 4.12E204) 0.11) 3.89E201)cg07437737) chr10) SVIL) MIR938) 12452) 0.11) 2.52E206) 0.10) 1.16E202)cg26188621) chr10) VDAC2) COMTD1) 12357) 0.13) 8.14E210) 0.24) 5.86E205)cg11451315) chr10) )) ANKRD30A) 2436731) 0.10) 2.15E206) 0.10) 1.07E202)cg24233594) chr10) )) AX747408) 243064) 0.11) 1.63E209) 0.13) 9.57E205)cg23931150) chr10) )) DKK1) 40184) 0.10) 8.45E207) 0.08) 5.22E203)cg14578311) chr10) )) GTPBP4) 214561) 0.14) 7.64E210) 0.00) 5.59E205)cg04869006) chr10) )) LOC644215) 7381) 0.11) 1.93E212) 0.01) 3.11E206)211  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg10593922) chr10) )) MAP3K8) 211567) 0.13) 2.85E208) 0.08) 5.91E204)cg00496126) chr10) )) NEBL) 2106707) 0.15) 2.76E209) 0.07) 1.29E204)cg26276667) chr10) )) NUDT9P1) 23452) 0.12) 1.25E208) 0.08) 3.02E204)cg05467828) chr10) )) PWWP2B) 23571) 0.10) 1.49E206) 0.11) 7.78E203)cg09623279) chr10) )) TMEM72) 211029) 0.11) 3.15E207) 0.11) 2.53E203)cg23330212) chr11) BDNFOS) BDNF) 8499) 0.11) 1.47E207) 0.11) 1.75E203)cg09781307) chr11) BDNFOS) EF565102) 7804) 0.10) 6.87E206) 0.08) 2.54E202)cg15172785) chr11) DYNC2H1) DYNC2H1) 125759) 0.11) 1.36E206) 0.06) 6.63E203)cg05521767) chr11) GDPD5) GDPD5) 6464) 0.17) 2.49E211) 0.18) 9.58E206)cg00510149) chr11) MRVI1) MRVI1) 21118) 0.11) 1.44E207) 0.09) 1.67E203)cg15433043) chr11) MYO7A) MYO7A) 21400) 0.12) 7.02E210) 0.17) 4.98E205)cg12892243) chr11) NPAT) NPAT) 211101) 0.12) 7.38E212) 0.18) 5.08E206)cg21660452) chr11) NRXN2) NRXN2) 26339) 0.12) 1.19E206) 0.06) 6.25E203)cg12387464) chr11) OR8D4) OR8D4) 21203) 0.10) 5.53E205) 0.01) 1.39E201)cg16243351) chr11) PAMR1) SLC1A2) 211980) 0.13) 1.93E209) 0.11) 1.07E204)cg07579831) chr11) SHANK2) SHANK2) 248682) 0.11) 1.26E205) 0.04) 4.11E202)cg13844463) chr11) SLC1A2) SLC1A2) 4433) 0.10) 3.81E207) 0.05) 3.08E203)cg21961694) chr11) SLC22A25) AY517500) 212885) 0.10) 7.80E205) 0.00) 1.72E201)cg20092728) chr11) SLC5A12) SLC5A12) 507) 0.12) 2.75E204) 0.07) 3.19E201)cg17284609) chr11) SOX6) SOX6) 2260) 0.11) 1.25E205) 0.03) 4.08E202)cg22061561) chr11) TMEM86A) TMEM86A) 3638) 0.12) 4.75E208) 0.09) 7.68E204)cg18247223) chr11) TSPAN18) TSPAN18) 219284) 0.10) 1.52E206) 0.07) 7.97E203)cg17425351) chr11) )) 5S_rRNA) 67887) 0.10) 3.67E206) 0.09) 1.55E202)cg07819010) chr11) )) AMBRA1) 23804) 0.13) 1.42E209) 0.01) 8.79E205)212  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg05394210) chr11) )) DCDC5) 21981) 0.12) 7.24E204) 0.08) 5.01E201)cg07530172) chr11) )) FAM76B) 91581) 0.10) 2.09E206) 0.11) 1.05E202)cg08716255) chr11) )) FAM76B) 89610) 0.10) 9.50E205) 0.02) 1.95E201)cg20733250) chr11) )) GRIK4) 25385) 0.11) 1.91E206) 0.04) 9.22E203)cg15824435) chr11) )) LOC100506540) 24144) 0.10) 6.86E203) 0.01) 9.42E201)cg10752604) chr11) )) LOC283143) 106494) 0.11) 8.99E205) 0.02) 1.89E201)cg26378984) chr11) )) Mir_652) 224329) 0.10) 5.73E208) 0.10) 8.84E204)cg18564881) chr11) )) PPFIA1) 225738) 0.13) 1.47E209) 0.04) 9.20E205)cg04942251) chr11) )) RCOR2) 22931) 0.12) 1.99E208) 0.08) 4.28E204)cg24735129) chr11) )) SHANK2) 221678) 0.11) 2.15E211) 0.04) 9.00E206)cg02058483) chr11) )) SLC1A2) 21308) 0.10) 1.76E205) 0.07) 5.51E202)cg20395789) chr11) )) SLN) 7775) 0.11) 1.58E209) 0.04) 9.37E205)cg04945434) chr11) )) SnoMBII_202) 2165105) 0.13) 4.30E207) 0.04) 3.54E203)cg23104436) chr11) )) SnoMBII_202) 2335308) 0.11) 6.96E207) 0.11) 4.48E203)cg19578175) chr11) )) SOX6) 459914) 0.13) 4.13E208) 0.05) 7.22E204)cg08198187) chr12) AMN1) AMN1) 2488) 0.12) 1.43E210) 0.14) 1.88E205)cg06495803) chr12) CD27) CD27) 2561) 0.11) 3.98E208) 0.02) 7.01E204)cg27306537) chr12) GPR109A) HCAR2) 375) 0.11) 1.05E208) 0.01) 2.82E204)cg25390063) chr12) HMGA) AK128707) 233405) 0.11) 9.88E206) 0.05) 3.25E202)cg09676622) chr12) NCOR2) NCOR2) 245644) 0.10) 1.51E210) 0.02) 1.90E205)cg21626573) chr12) NCOR2) NCOR2) 245573) 0.14) 7.10E213) 0.15) 1.78E206)cg04708790) chr12) OAS1) OAS1) 10937) 0.11) 2.33E205) 0.01) 6.63E202)cg15534461) chr12) PEBP1) PEBP1) 1164) 0.11) 2.83E208) 0.07) 5.52E204)cg05399718) chr12) PITPNM2) LOC100507091) 210395) 0.10) 3.14E209) 0.01) 1.39E204)213  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg20389635) chr12) PTHLH) PTHLH) 2840) 0.11) 4.74E206) 0.03) 1.84E202)cg15282973) chr12) SCARB1) SCARB1) 2266) 0.16) 1.37E211) 0.12) 7.53E206)cg04950931) chr12) TSPAN8) TSPAN8) 2678) 0.11) 6.00E206) 0.06) 2.41E202)cg08338281) chr12) WIBG) WIBG) 2970) 0.12) 1.81E204) 0.07) 2.57E201)cg09676376) chr12) ZNF385A) ZNF385A) 475) 0.13) 5.66E211) 0.03) 1.40E205)cg12150931) chr12) ZNF385A) ZNF385A) 2301) 0.11) 3.06E210) 0.03) 2.84E205)cg26851803) chr12) )) 5S_rRNA) 267346) 0.15) 6.31E210) 0.00) 4.37E205)cg16964478) chr12) )) BC041342) 21228) 0.12) 2.92E205) 0.12) 7.45E202)cg12647920) chr12) )) CORO1C) 219449) 0.16) 2.08E208) 0.13) 4.51E204)cg18689240) chr12) )) KRT7) 22214) 0.11) 1.79E204) 0.02) 2.55E201)cg23128495) chr12) )) NUAK1) 279328) 0.12) 3.75E206) 0.05) 1.56E202)cg12493761) chr12) )) PRICKLE1) 2269685) 0.11) 2.66E209) 0.11) 1.25E204)cg25103160) chr12) )) SNORA70) 2123739) 0.11) 2.55E209) 0.01) 1.19E204)cg06377626) chr13) BIVM) KDELC1) 22772) 0.11) 6.66E210) 0.01) 4.62E205)cg01427300) chr13) CAB39L) CAB39L) 53856) 0.10) 7.06E208) 0.06) 1.12E203)cg15887927) chr13) )) DD413682) 225138) 0.15) 7.31E212) 0.04) 4.89E206)cg24561305) chr13) )) DD413682) 224958) 0.12) 1.96E212) 0.11) 3.25E206)cg03099780) chr13) )) MBNL2) 210098) 0.14) 1.36E211) 0.08) 6.96E206)cg14920808) chr13) )) PABPC3) 20554) 0.13) 2.33E207) 0.08) 2.24E203)cg26651514) chr13) )) STARD13) 24833) 0.20) 1.05E211) 0.11) 5.49E206)cg19291696) chr13) )) USP12) 223404) 0.10) 4.47E206) 0.10) 1.82E202)cg05131266) chr14)C14orf159;SNORA11B) SNORA11B) 2881) 0.15) 1.32E212) 0.05) 2.49E206)214  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg07527324) chr14)C14orf159;SNORA11B) SNORA11B) 21150) 0.11) 1.09E210) 0.05) 1.82E205)cg21906866) chr14)C14orf159;SNORA11B) SNORA11B) 290) 0.12) 1.01E207) 0.17) 1.40E203)cg12140749) chr14) C14orf181) C14orf181) 21028) 0.11) 1.18E209) 0.04) 7.67E205)cg12156272) chr14) C14orf184) C14orf184) 2873) 0.11) 2.59E205) 0.07) 7.02E202)cg03539735) chr14) EML1) EVL) 235492) 0.10) 1.79E209) 0.08) 1.01E204)cg22800400) chr14) FRMD6) FRMD6) 2330) 0.10) 3.11E210) 0.02) 2.94E205)cg02275040) chr14) FRMD6) U6) 17593) 0.11) 4.21E208) 0.01) 7.42E204)cg14377923) chr14) MTA1) MTA1) 2941) 0.10) 1.19E207) 0.12) 1.48E203)cg18114811) chr14) TCL6) TCL6) 2135) 0.11) 1.76E207) 0.13) 1.92E203)cg23696550) chr14) TGM1) TGM1) 30) 0.14) 4.28E211) 0.04) 1.19E205)cg06617636) chr14) ZFP36L1) ZFP36L1) 3941) 0.11) 1.22E205) 0.03) 4.06E202)cg07095945) chr15) ALPK3) ALPK3) 7697) 0.11) 1.68E208) 0.11) 3.82E204)cg02633229) chr15) C15orf23) C15orf23) 243) 0.12) 1.53E207) 0.05) 1.79E203)cg12732548) chr15) C15orf52) C15orf52) 21239) 0.11) 3.08E208) 0.04) 6.21E204)cg03728457) chr15) CCDC33) CCDC33) 8696) 0.18) 1.72E211) 0.00) 8.58E206)cg06285340) chr15) CYP11A1) CYP11A1) 180) 0.10) 8.38E209) 0.13) 2.48E204)cg20330630) chr15) CYP11A1) CYP11A1) 2294) 0.12) 3.26E208) 0.06) 6.29E204)cg18512205) chr15) GPR176) GPR176) 29534) 0.12) 5.35E210) 0.08) 4.18E205)cg19674091) chr15) IDH2) IDH2) 161) 0.15) 9.73E207) 0.02) 5.93E203)cg14851263) chr15) MYO5A) MYO5A) 21335) 0.10) 2.71E204) 0.02) 3.13E201)cg22234930) chr15) PKM2) PKM2) 1278) 0.21) 2.65E212) 0.07) 3.43E206)215  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg20315739) chr15) PLA2G4E) PLA2G4E) 213312) 0.11) 1.31E205) 0.02) 4.24E202)cg18120790) chr15) TRPM1) MTMR10) 211251) 0.10) 2.81E205) 0.06) 7.41E202)cg19786983) chr15) )) AK094929) 68936) 0.12) 3.02E208) 0.18) 6.15E204)cg17383853) chr15) )) AK125516) 238785) 0.12) 3.47E207) 0.10) 2.85E203)cg12290217) chr15) )) AL109706) 28742) 0.12) 2.04E210) 0.14) 2.21E205)cg17357548) chr15) )) ATP8B4) 218410) 0.10) 2.18E207) 0.04) 2.15E203)cg25653839) chr15) )) BUB1B) 23183) 0.12) 4.07E208) 0.06) 7.17E204)cg26861706) chr15) )) CHSY1) 41736) 0.11) 4.45E209) 0.15) 1.75E204)cg21733367) chr15) )) DQ575076) 2428) 0.11) 5.74E205) 0.04) 1.42E201)cg25175240) chr15) )) LOC145837) 24272) 0.13) 9.52E207) 0.06) 5.85E203)cg25130939) chr15) )) NRG4) 223355) 0.11) 8.31E206) 0.05) 2.94E202)cg10548163) chr15) )) Y_RNA) 49762) 0.13) 3.61E207) 0.14) 3.03E203)cg09648454) chr16) ABCC12) ABCC12) 22825) 0.10) 2.81E206) 0.08) 1.31E202)cg07762788) chr16) CDH1) TMCO7) 244202) 0.12) 3.68E207) 0.08) 3.05E203)cg01033642) chr16) CMIP) CMIP) 35239) 0.13) 1.17E214) 0.08) 2.50E207)cg02547035) chr16) CMIP) CMIP) 35600) 0.12) 2.17E210) 0.04) 2.49E205)cg04897892) chr16) CMIP) CMIP) 35361) 0.16) 6.05E211) 0.09) 1.41E205)cg10246581) chr16) CMIP) CMIP) 7218) 0.18) 1.75E213) 0.17) 6.39E207)cg16353318) chr16) CMIP) CMIP) 35208) 0.13) 6.95E216) 0.09) 1.96E207)cg03777414) chr16) FAM18A) FAM18A) 221) 0.20) 7.36E211) 0.13) 1.47E205)cg07860918) chr16) GAS8) C16orf3) 3554) 0.12) 1.89E206) 0.04) 9.17E203)cg02843417) chr16) IRF8) IRF8) 13462) 0.10) 8.46E212) 0.07) 5.40E206)cg06665333) chr16) SLC7A5) SLC7A5) 29263) 0.12) 9.34E206) 0.01) 3.19E202)cg09038266) chr16) WWOX) MAF) 478664) 0.12) 8.38E206) 0.08) 2.97E202)216  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg00902849) chr16) )) AK093002) 2290) 0.10) 4.07E204) 0.09) 3.89E201)cg16468174) chr16) )) BC013599) 4809) 0.10) 6.17E209) 0.03) 2.15E204)cg01323840) chr16) )) CPPED1) 23364) 0.12) 1.64E208) 0.04) 3.72E204)cg16804825) chr16) )) KIAA0182) 250430) 0.12) 1.20E212) 0.01) 2.48E206)cg07443080) chr16) )) LINC00311) 210672) 0.11) 3.04E203) 0.01) 7.95E201)cg04413904) chr16) )) SSTR5) 234276) 0.12) 1.07E205) 0.12) 3.58E202)cg04696808) chr16) )) UBE2MP1) 147592) 20.12) 2.31E203) 20.01) 7.30E201)cg02441982) chr17) AFMID) AFMID) 244) 0.13) 8.31E209) 0.09) 2.44E204)cg21823502) chr17) ANAPC11) NPB) 23641) 0.12) 2.09E210) 0.10) 2.32E205)cg04662250) chr17) C17orf103) C17orf103) 2921) 0.11) 1.73E209) 0.12) 9.94E205)cg19760250) chr17) C17orf98) C17orf98) 15) 0.11) 4.03E206) 0.05) 1.65E202)cg26548883) chr17)CCL142CCL15) CCL142CCL15) 2815) 0.10) 1.05E205) 0.09) 3.50E202)cg13724311) chr17) CLDN7) CLDN7) 2665) 0.11) 7.91E212) 0.19) 5.38E206)cg02023717) chr17) ERBB2) ERBB2) 2305) 0.11) 1.39E207) 0.11) 1.62E203)cg12779575) chr17) ERN1) ERN1) 2932) 0.18) 1.58E212) 0.27) 2.92E206)cg01257345) chr17)IFT20;TMEM97) TNFAIP1) 22590) 0.14) 3.93E208) 0.03) 6.74E204)cg20438445) chr17) KCNH4) RAB5C) 22295) 0.11) 6.33E209) 0.07) 2.20E204)cg17464043) chr17) KRT39) KRT39) 7693) 0.15) 1.63E210) 0.04) 2.01E205)cg22303227) chr17) LOC90586) AOC4) 2169) 0.11) 1.39E208) 0.10) 3.41E204)cg24569276) chr17) PIGS) PIGS) 8015) 0.11) 1.91E213) 0.15) 7.18E207)cg15432510) chr17) RPTOR) AK127919) 5343) 0.12) 7.96E207) 0.06) 4.78E203)cg03720572) chr17) SLC43A2) SLC43A2) 18540) 0.10) 1.42E204) 0.00) 2.34E201)cg09822726) chr17) TANC2) TANC2) 11188) 0.11) 1.43E204) 0.04) 2.36E201)217  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg12961926) chr17) TBCD) TBCD) 214400) 0.13) 1.98E208) 0.01) 4.26E204)cg14605961) chr17) TBCD) TBCD) 212107) 0.12) 1.56E206) 0.00) 8.24E203)cg16518729) chr17) TBCD) TBCD) 28125) 0.15) 5.47E206) 0.11) 2.24E202)cg24481782) chr17) TBCD) TBCD) 28047) 0.15) 1.12E207) 0.01) 1.46E203)cg07564598) chr17) WSB1) MIR4522) 1076) 0.12) 9.95E207) 0.07) 6.08E203)cg07168214) chr17) ZBTB4) ZBTB4) 2832) 0.13) 2.30E211) 0.14) 9.54E206)cg03835038) chr17) )) AB062083) 6583) 0.10) 1.10E204) 0.01) 2.13E201)cg15082983) chr17) )) HAP1) 23882) 0.11) 2.01E205) 0.08) 6.26E202)cg26625897) chr17) )) KRT15) 2653) 0.29) 1.02E215) 0.09) 2.00E207)cg09976845) chr17) )) MED13) 269387) 0.12) 2.04E209) 0.07) 1.09E204)cg08805241) chr17) )) RNF126P1) 211934) 0.10) 3.75E206) 0.07) 1.55E202)cg20090782) chr18) SETBP1) SETBP1) 62854) 0.12) 9.05E211) 0.18) 1.69E205)cg22035959) chr19) ARID3A) ARID3A) 25645) 0.10) 1.32E211) 0.13) 6.49E206)cg25298189) chr19) ARID3A) ARID3A) 9223) 0.10) 5.89E208) 0.06) 9.55E204)cg22981158) chr19) DACT3) LOC100506068) 27219) 0.12) 1.37E206) 0.08) 6.87E203)cg03752885) chr19) DAPK3) DAPK3) 91) 0.14) 1.52E209) 0.05) 9.22E205)cg26585416) chr19) DAPK3) DAPK3) 82) 0.14) 3.25E207) 0.14) 2.76E203)cg09412728) chr19) HIPK4) HIPK4) 21373) 0.11) 2.61E206) 0.09) 1.18E202)cg05213896) chr19) IL4I1) MIR4750) 2226) 0.11) 1.06E205) 0.05) 3.51E202)cg07381806) chr19) MOBKL2A) MOB3A) 1942) 0.13) 1.45E206) 0.06) 7.74E203)cg24938761) chr19) MOBKL2A) MOB3A) 4863) 0.10) 2.98E207) 0.09) 2.43E203)cg11387248) chr19) PSG8) PSG8) 246) 0.12) 1.45E209) 0.09) 9.02E205)cg24062310) chr19) RAB4B) RAB4B) 5043) 0.11) 1.92E205) 0.03) 6.10E202)cg18743287) chr19) RDH13) AK122764) 2636) 0.15) 4.60E208) 0.05) 7.54E204)218  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg02821871) chr19) SPHK2) SPHK2) 4451) 0.10) 1.02E206) 0.13) 6.08E203)cg04981492) chr19) SYDE1) SYDE1) 500) 0.14) 6.09E213) 0.05) 1.51E206)cg18584265) chr19) SYDE1) SYDE1) 340) 0.15) 2.80E213) 0.14) 1.29E206)cg14584702) chr19) USHBP1) USHBP1) 2928) 0.10) 2.91E206) 0.06) 1.33E202)cg10668363) chr19) ZNF175) ZNF175) 2161) 0.17) 8.79E209) 0.03) 2.60E204)cg25722029) chr19) ZSWIM4) ZSWIM4) 1423) 0.11) 7.77E209) 0.07) 2.36E204)cg22526990) chr19) )) CEBPA) 4919) 0.13) 2.37E213) 0.16) 9.46E207)cg00713022) chr19) )) SYDE1) 22270) 0.15) 4.87E207) 0.09) 3.67E203)cg10455413) chr2) AGAP1) AGAP1) 49872) 0.10) 1.16E209) 0.04) 7.45E205)cg12833765) chr2) ANTXR1) MIR3126) 140065) 0.13) 8.08E208) 0.11) 1.20E203)cg16980530) chr2) BMP10) BMP10) 2935) 0.13) 7.32E205) 0.06) 1.64E201)cg06867755) chr2) CUL3) CUL3) 37155) 0.11) 7.89E209) 0.05) 2.41E204)cg22245446) chr2) DIRC3) DIRC3) 149101) 0.11) 5.77E206) 0.07) 2.33E202)cg15129144) chr2) EPAS1) EPAS1) 3418) 0.13) 5.70E210) 0.01) 4.19E205)cg03506656) chr2) FAP) FAP) 2535) 0.11) 1.82E205) 0.06) 5.65E202)cg13754437) chr2) FHL2) FHL2) 14) 0.21) 1.11E212) 0.09) 2.40E206)cg11348701) chr2) GAD1) GAD1) 21353) 0.10) 5.27E206) 0.01) 2.10E202)cg22860917) chr2) GALM) GALM) 26300) 0.14) 1.82E208) 0.14) 4.06E204)cg07591442) chr2) GFPT1) GFPT1) 2719) 0.10) 1.37E204) 0.03) 2.28E201)cg12157646) chr2) GPR39) GPR39) 15553) 0.11) 2.16E205) 0.07) 6.43E202)cg24964130) chr2) GPR39) GPR39) 23431) 0.11) 1.43E206) 0.11) 7.32E203)cg22330749) chr2) GRHL1) BC051708) 227706) 0.11) 4.90E208) 0.11) 8.04E204)cg24617203) chr2) IL1R2) IL1R2) 21595) 0.15) 1.13E210) 0.02) 1.84E205)cg22827324) chr2) LTBP1) RASGRP3) 248539) 0.14) 2.49E209) 0.04) 1.15E204)219  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg12686754) chr2) MIR216A) MIR216A) 2389) 0.11) 3.65E208) 0.04) 6.51E204)cg21182526) chr2)MIR548N;OSBPL6) MIR548N) 13978) 0.11) 2.63E207) 0.11) 2.35E203)cg24882192) chr2) MYO7B) MYO7B) 211232) 0.14) 2.13E208) 0.04) 4.55E204)cg09367174) chr2) NEB) NEB) 2165) 0.12) 2.38E207) 0.08) 2.26E203)cg24430754) chr2) NMUR1) NMUR1) 21059) 0.11) 9.73E208) 0.07) 1.35E203)cg15005526) chr2) NOTO) SMYD5) 23223) 0.11) 1.63E207) 0.02) 1.83E203)cg08900316) chr2) PRKCE) EPAS1) 2223457) 0.14) 1.16E210) 0.20) 1.86E205)cg23101885) chr2)RAD51AP2) RAD51AP2) 3681) 0.13) 2.53E208) 0.09) 5.37E204)cg03180819) chr2) SCN3A) SCN3A) 21022) 0.12) 4.21E212) 0.05) 3.92E206)cg22344841) chr2) SDC1) SDC1) 1493) 0.13) 1.66E211) 0.14) 8.02E206)cg03339817) chr2) SFT2D3) SFT2D3) 2315) 0.16) 8.49E209) 0.11) 2.49E204)cg11206634) chr2) SFT2D3) SFT2D3) 2197) 0.10) 5.45E208) 0.01) 8.76E204)cg24925163) chr2) SFT2D3) SFT2D3) 2348) 0.15) 3.05E209) 0.04) 1.39E204)cg09500437) chr2) SH3RF3) MIR4265) 255132) 0.10) 4.20E208) 0.12) 7.37E204)cg17010118) chr2) SLC11A1) SLC11A1) 4522) 0.14) 1.73E209) 0.02) 9.76E205)cg20822693) chr2) SLC16A14) SLC16A14) 2684) 0.11) 6.23E208) 0.02) 9.71E204)cg24995347) chr2) SP140) SP110) 803) 0.11) 3.02E210) 0.02) 2.81E205)cg07903472) chr2) STK39) B3GALT1) 121108) 0.11) 6.50E205) 0.06) 1.52E201)cg13345122) chr2) STK39) B3GALT1) 187336) 0.11) 4.22E205) 0.06) 1.08E201)cg18947276) chr2)STON12GTF2A1L)STON12GTF2A1L) 21091) 0.11) 3.71E210) 0.15) 3.32E205)cg21560697) chr2) TAF1B) GRHL1) 236908) 0.12) 1.01E209) 0.07) 6.69E205)cg02286857) chr2) TTC7A) TTC7A) 75808) 0.10) 1.83E208) 0.07) 4.15E204)220  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg26638266) chr2) VIT) VIT) 2746) 0.11) 6.84E213) 0.15) 1.52E206)cg16326590) chr2) VWA3B) VWA3B) 1264) 0.12) 2.44E210) 0.05) 2.56E205)cg09638264) chr2) WDR69) WDR69) 2604) 0.12) 4.28E207) 0.02) 3.51E203)cg08250187) chr2) )) 7SK) 2178048) 0.12) 8.67E207) 0.08) 5.67E203)cg20497304) chr2) )) ATP5G3) 248827) 0.12) 2.01E206) 0.08) 9.97E203)cg17933206) chr2) )) AX748264) 214334) 0.11) 2.61E208) 0.03) 5.47E204)cg01491071) chr2) )) BC047484) 2160539) 0.10) 4.39E207) 0.10) 3.56E203)cg10440877) chr2) )) CREB1) 216140) 0.10) 3.46E209) 0.02) 1.51E204)cg02095697) chr2) )) FAM150B) 2109422) 0.10) 2.07E204) 0.06) 2.77E201)cg12436772) chr2) )) FN1) 2101593) 0.27) 2.00E212) 0.15) 3.38E206)cg02625481) chr2) )) FRZB) 64374) 0.11) 3.80E208) 0.03) 6.74E204)cg11945095) chr2) )) GRHL1) 24868) 0.12) 1.59E210) 0.03) 2.00E205)cg06375062) chr2) )) LOC100128590) 4485) 0.10) 8.65E207) 0.04) 5.65E203)cg01413880) chr2) )) Mir_548) 28732) 0.13) 2.77E211) 0.06) 9.62E206)cg18918418) chr2) )) Mir_548) 1828) 0.14) 1.95E206) 0.08) 9.61E203)cg27492725) chr2) )) PDE1A) 221964) 0.10) 3.22E207) 0.05) 2.67E203)cg02287004) chr2) )) RDH14) 135479) 0.10) 3.22E206) 0.06) 1.41E202)cg27223882) chr2) )) RPRM) 89976) 0.11) 2.04E204) 0.01) 2.71E201)cg18199208) chr2) )) SLC4A3) 23005) 0.11) 6.84E213) 0.10) 1.57E206)cg05074732) chr2) )) STK39) 292019) 0.11) 1.15E205) 0.04) 3.70E202)cg24761668) chr2) )) SULT1C4) 31701) 0.10) 1.78E203) 0.08) 6.80E201)cg20600404) chr2) )) TAF1B) 222987) 0.11) 1.86E208) 0.02) 4.22E204)cg23255934) chr2) )) TRNA_Pseudo) 2194088) 0.10) 9.13E205) 0.08) 1.90E201)cg14938029) chr20) CSTL1) CSTL1) 967) 0.11) 2.07E205) 0.08) 6.34E202)221  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg16606561) chr20) FAM110A) FAM110A) 2643) 0.17) 2.65E213) 0.01) 9.85E207)cg26331247) chr20)LOC284805) C20orf203) 612) 0.10) 9.92E207) 0.09) 6.00E203)cg00990977) chr20) RRBP1) RRBP1) 3611) 0.14) 3.15E208) 0.09) 6.22E204)cg12632411) chr20) RRBP1) RRBP1) 5107) 0.17) 5.68E213) 0.03) 1.47E206)cg24023487) chr20) SPTLC3) SPTLC3) 2899) 0.13) 2.24E206) 0.12) 1.09E202)cg15799353) chr20) TRIB3) TRIB3) 1172) 0.13) 3.62E213) 0.08) 1.31E206)cg15582126) chr20) )) BTBD3) 23969) 0.11) 5.93E205) 0.03) 1.45E201)cg10187713) chr20) )) PARD6B) 22393) 0.19) 1.41E211) 0.05) 7.58E206)cg14867395) chr21) BACE2) PLAC4) 21489) 0.12) 1.61E206) 0.09) 8.50E203)cg08875503) chr21) BTG3) BTG3) 4960) 0.12) 1.14E207) 0.08) 1.47E203)cg25032603) chr21) C21orf82) LINC00310) 1834) 0.16) 2.10E210) 0.12) 2.37E205)cg02475236) chr21) DSCAM) DSCAM) 2038) 0.12) 5.38E206) 0.05) 2.17E202)cg14746283) chr21) MX1) pp9284) 8336) 0.12) 2.32E205) 0.07) 6.58E202)cg19118951) chr21) )) LINC00310) 22093) 0.12) 1.75E208) 0.13) 3.92E204)cg08612539) chr22) NCF4) NCF4) 95) 0.14) 3.72E212) 0.07) 3.80E206)cg20478080) chr22) PRR5) PRR5) 6483) 0.11) 6.83E210) 0.02) 4.89E205)cg25245338) chr22)SYN3;TIMP3) TIMP3) 2689) 0.19) 6.14E210) 0.02) 4.26E205)cg09476306) chr22) YPEL1) YPEL1) 3736) 0.10) 8.38E207) 0.02) 5.21E203)cg20080983) chr22) ZNRF3) C22orf31) 27636) 0.12) 1.10E209) 0.10) 7.33E205)cg20012247) chr22) )) DGCR8) 23213) 0.14) 4.19E207) 0.05) 3.48E203)cg05663031) chr3) BCL6) BCL6) 564) 0.12) 1.21E211) 0.18) 6.22E206)cg06070445) chr3) BCL6) BCL6) 513) 0.12) 1.55E209) 0.16) 9.25E205)cg17394304) chr3) BCL6) BCL6) 312) 0.14) 7.27E214) 0.05) 6.32E207)222  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg16582517) chr3) BHLHE40) LOC100507582) 24239) 0.13) 1.59E206) 0.09) 8.47E203)cg26269881) chr3) BHLHE40) LOC100507582) 21664) 0.10) 2.58E208) 0.04) 5.39E204)cg14454680) chr3)C3orf26;FILIP1L) FILIP1L) 39339) 20.10) 3.14E204) 20.04) 3.48E201)cg20970886) chr3) C3orf59) MB21D2) 67236) 0.13) 4.55E210) 0.05) 4.01E205)cg10144104) chr3)CACNA2D2) CACNA2D2) 21996) 0.11) 5.31E206) 0.07) 2.14E202)cg01519765) chr3) ETV5) ETV5) 21219) 0.13) 1.15E208) 0.03) 2.90E204)cg02026180) chr3) FLNB) FLNB) 34471) 0.11) 7.44E207) 0.11) 4.56E203)cg23730027) chr3) FLNB) FLNB) 1054) 0.17) 8.79E211) 0.05) 1.69E205)cg24204017) chr3) FLNB) FLNB) 5862) 0.14) 4.68E212) 0.09) 4.34E206)cg15318697) chr3) GPD1L) GPD1L) 39899) 0.11) 1.70E205) 0.06) 5.41E202)cg18863595) chr3) GRM7) GRM7) 44) 20.11) 3.74E204) 20.04) 3.78E201)cg03349057) chr3) ITIH4) ITIH4) 5151) 0.11) 3.30E206) 0.04) 1.42E202)cg20570458) chr3) KALRN) KALRN) 233738) 0.11) 6.48E207) 0.03) 4.42E203)cg19937056) chr3) KCNMB2) KCNMB2) 2919) 0.11) 2.91E204) 0.02) 3.34E201)cg19234171) chr3) LOC90246) LOC90246) 2606) 0.12) 2.15E208) 0.08) 4.56E204)cg13672136) chr3) LTF) LTF) 27480) 0.13) 2.58E209) 0.12) 1.24E204)cg27502580) chr3) MECOM) DD413694) 132454) 0.11) 4.18E206) 0.07) 1.68E202)cg10638793) chr3) MECOM) MECOM) 71708) 0.10) 5.24E207) 0.04) 3.83E203)cg08723913) chr3) MIR568) MIR568) 2534) 0.11) 5.58E204) 0.00) 4.53E201)cg03851970) chr3) MITF) MITF) 6562) 0.11) 5.29E205) 0.01) 1.28E201)cg03387092) chr3) MYLK) MYLK) 19443) 0.13) 9.69E206) 0.03) 3.23E202)cg06056170) chr3) NFKBIZ) NFKBIZ) 22420) 0.10) 5.10E209) 0.03) 1.94E204)223  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg20782816) chr3)OGG1;CAMK1) CAMK1) 3608) 0.11) 1.45E207) 0.03) 1.70E203)cg26084141) chr3) PDIA5) PDIA5) 1040) 0.12) 3.00E207) 0.12) 2.43E203)cg01412654) chr3) PPARG) PPARG) 21264) 0.13) 1.30E209) 0.05) 8.65E205)cg18751231) chr3) SCHIP1) U2) 21319) 0.13) 5.53E206) 0.05) 2.27E202)cg05507514) chr3) SGEF) ARHGEF26) 3698) 0.11) 3.17E207) 0.08) 2.56E203)cg04858987) chr3) SH3BP5) SH3BP5) 14766) 0.14) 1.05E207) 0.15) 1.40E203)cg18444702) chr3) SH3BP5) SH3BP5) 7242) 0.12) 2.73E205) 0.05) 7.25E202)cg05937737) chr3) SLC7A14) SLC7A14) 323) 20.11) 3.20E204) 20.03) 3.50E201)cg16275903) chr3) SRPRB) SRPRB) 21130) 0.10) 5.90E210) 0.03) 4.21E205)cg06520450) chr3) TP63) TP63) 252) 0.10) 1.20E205) 0.04) 3.92E202)cg06641593) chr3) VILL) VILL) 2842) 0.11) 4.95E209) 0.09) 1.87E204)cg01127412) chr3) XIRP1) XIRP1) 1228) 0.11) 3.11E209) 0.09) 1.39E204)cg02147465) chr3) ZBTB20) ZBTB202AS1) 22518) 0.11) 3.68E204) 0.00) 3.70E201)cg09624551) chr3) )) AK097190) 21192) 0.11) 1.71E207) 0.03) 1.91E203)cg08678314) chr3) )) ARHGEF26) 2159494) 0.11) 1.36E205) 0.09) 4.51E202)cg11624780) chr3) )) AX747368) 2732) 0.11) 1.41E207) 0.00) 1.65E203)cg16291750) chr3) )) BC035826) 2502) 0.11) 4.47E204) 0.04) 4.07E201)cg13546308) chr3) )) C3orf79) 163872) 0.14) 9.78E208) 0.05) 1.37E203)cg00445523) chr3) )) CACNA2D2) 219564) 0.10) 5.24E207) 0.08) 3.93E203)cg13062627) chr3) )) CSRNP1) 212261) 0.15) 4.21E208) 0.07) 7.42E204)cg12979789) chr3) )) FAM19A1) 25518) 0.10) 1.08E204) 0.03) 2.10E201)cg02770406) chr3) )) FLNB) 22676) 0.20) 3.73E210) 0.12) 3.38E205)cg22305268) chr3) )) HES1) 3610) 0.15) 5.79E207) 0.10) 4.24E203)cg18813778) chr3) )) KCNH8) 273686) 0.11) 1.09E205) 0.07) 3.60E202)224  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg21760128) chr3) )) LARS2) 22471) 0.10) 8.63E208) 0.08) 1.25E203)cg27113538) chr3) )) LOC401109) 219832) 0.12) 7.01E208) 0.03) 1.11E203)cg06340382) chr3) )) P2RY1) 288610) 0.12) 1.01E205) 0.06) 3.32E202)cg15765546) chr3) )) SLCO2A1) 227437) 0.14) 5.72E212) 0.11) 4.47E206)cg01571842) chr3) )) SLMAP) 52636) 0.11) 1.25E208) 0.11) 2.99E204)cg21977522) chr3) )) U6atac) 20133) 0.13) 1.21E204) 0.07) 2.23E201)cg13785473) chr4) ADD1) MFSD10) 5187) 0.11) 1.40E206) 0.07) 6.92E203)cg11474135) chr4) ANK2) ANK2) 237483) 0.10) 3.45E205) 0.06) 8.76E202)cg16253629) chr4) ANK2) ANK2) 36978) 0.11) 9.91E205) 0.01) 2.00E201)cg25429446) chr4) ANK2) ANK2) 99807) 0.11) 2.66E206) 0.07) 1.28E202)cg08116915) chr4) CPZ) CPZ) 2662) 0.11) 1.51E208) 0.05) 3.47E204)cg04038163) chr4) CXCL9) CXCL9) 3486) 0.14) 8.67E211) 0.11) 1.67E205)cg11637968) chr4) FNIP2) FNIP2) 41855) 0.15) 2.54E209) 0.15) 1.19E204)cg12982893) chr4) GAB1) GAB1) 14734) 0.10) 1.55E205) 0.05) 5.20E202)cg23160016) chr4) GABRA2) GABRA2) 127) 20.13) 1.38E205) 20.08) 4.65E202)cg16177314) chr4) GYPB) GYPB) 21113) 0.11) 3.23E207) 0.08) 2.70E203)cg08112737) chr4) KIAA1211) KIAA1211) 2924) 0.12) 2.26E211) 0.10) 9.51E206)cg03174768) chr4) LIMCH1) LIMCH1) 7106) 0.10) 2.98E209) 0.12) 1.37E204)cg03822934) chr4) LIMCH1) LIMCH1) 234226) 0.16) 3.53E212) 0.00) 3.57E206)cg22204103) chr4) LIMCH1) LIMCH1) 275) 0.11) 2.46E206) 0.03) 1.13E202)cg25740652) chr4) LIMCH1) LIMCH1) 21180) 0.16) 2.16E208) 0.17) 4.60E204)cg22520372) chr4) NPY5R) NPY5R) 67) 0.11) 1.04E204) 0.07) 2.04E201)cg12845923) chr4) PDGFRA) PDGFRA) 23469) 0.11) 6.82E207) 0.09) 4.48E203)cg27442676) chr4) SLC4A4) SLC4A4) 17665) 0.11) 1.99E205) 0.02) 6.20E202)225  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg24715412) chr4) STOX2) STOX2) 27739) 0.11) 4.38E205) 0.10) 1.09E201)cg00813886) chr4) TBC1D14) TBC1D14) 214734) 0.10) 5.07E206) 0.02) 2.06E202)cg20054786) chr4) TLR1) TLR1) 1046) 0.10) 1.06E207) 0.02) 1.42E203)cg04597985) chr4) )) BC031238) 2203) 0.11) 1.84E203) 0.01) 6.83E201)cg08018143) chr4) )) BC033991) 2833130) 0.11) 6.71E204) 0.04) 4.77E201)cg19183252) chr4) )) BC038750) 59729) 0.11) 3.17E206) 0.01) 1.40E202)cg19749538) chr4) )) CCNG2) 35418) 0.13) 1.91E204) 0.10) 2.62E201)cg18579761) chr4) )) HNRPDL) 27793) 0.12) 3.01E207) 0.10) 2.44E203)cg10322735) chr4) )) LINC00290) 2349) 0.10) 2.11E203) 0.07) 7.07E201)cg06220112) chr4) )) LOC100505875) 111386) 0.14) 2.98E206) 0.02) 1.38E202)cg03078141) chr4) )) PHF17) 28929) 0.11) 1.10E207) 0.07) 1.42E203)cg26723624) chr4) )) RASL11B) 227554) 0.10) 5.95E211) 0.13) 1.40E205)cg06691616) chr4) )) SLC7A11) 2367595) 0.12) 6.57E206) 0.02) 2.49E202)cg01238639) chr4) )) TRAM1L1) 381987) 0.11) 2.96E205) 0.06) 7.74E202)cg02304751) chr4) )) TSPAN5) 24133) 0.10) 1.86E205) 0.06) 5.68E202)cg23831864) chr4) )) U6) 2282236) 0.13) 2.64E206) 0.05) 1.25E202)cg04543510) chr5) AGXT2) AGXT2) 22416) 0.10) 1.45E209) 0.05) 9.08E205)cg18190824) chr5) ANKH) FAM105B) 56244) 0.16) 1.01E211) 0.03) 5.47E206)cg07233933) chr5) CPEB4) CPEB4) 219269) 0.10) 4.86E208) 0.06) 8.04E204)cg07284261) chr5) CSF1R) CSF1R) 299) 0.13) 1.64E206) 0.04) 8.86E203)cg23309843) chr5) CTXN3) CTXN3) 2881) 0.11) 6.80E206) 0.06) 2.52E202)cg24598187) chr5) ERAP2) LNPEP) 217547) 0.13) 5.71E208) 0.03) 8.83E204)cg22793871) chr5) FAM151B) FAM151B) 2519) 20.13) 5.05E205) 20.03) 1.24E201)cg06609646) chr5) GRAMD3) GRAMD3) 2421) 0.10) 1.77E208) 0.19) 4.00E204)226  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg12825070) chr5) HTR4) HTR4) 31) 20.11) 8.18E205) 20.03) 1.77E201)cg06912936) chr5) KCNIP1) KCNIP1) 4150) 0.10) 5.19E211) 0.03) 1.28E205)cg13301676) chr5) MCC) MCC) 124452) 0.11) 6.03E205) 0.00) 1.47E201)cg06685875) chr5) PCSK1) PCSK1) 4668) 0.10) 1.55E206) 0.04) 8.10E203)cg15410418) chr5) PDE4D) PDE4D) 2794) 0.11) 1.02E208) 0.09) 2.68E204)cg18514949) chr5) PPP2CA) PPP2CA) 21266) 0.11) 1.07E206) 0.04) 6.09E203)cg08991927) chr5) PPP2R2B) PPP2R2B) 21105) 0.10) 5.13E205) 0.09) 1.25E201)cg19683073) chr5) SERINC5) SERINC5) 21369) 0.10) 6.89E205) 0.07) 1.60E201)cg21050001) chr5) SLC1A3) SLC1A3) 54374) 0.12) 2.98E205) 0.07) 7.94E202)cg15003812) chr5) TTC23L) RAD1) 16456) 0.11) 1.49E207) 0.14) 1.76E203)cg12508655) chr5) ZNF366) ZNF366) 2482) 0.12) 2.25E207) 0.04) 2.20E203)cg18259253) chr5) ZSWIM6) ZSWIM6) 117887) 0.10) 3.21E204) 0.00) 3.54E201)cg04677123) chr5) )) AK091866) 143718) 0.15) 3.49E210) 0.09) 3.09E205)cg05211068) chr5) )) AK091866) 35078) 0.10) 2.16E206) 0.08) 1.07E202)cg13931873) chr5) )) AK307154) 2829) 0.10) 3.41E206) 0.08) 1.51E202)cg16178353) chr5) )) ATP6AP1L) 29514) 0.10) 3.12E205) 0.03) 8.35E202)cg04577249) chr5) )) DUSP1) 18532) 0.13) 9.28E211) 0.05) 1.70E205)cg10472840) chr5) )) ESM1) 75353) 0.10) 8.70E205) 0.09) 1.87E201)cg05199904) chr5) )) MAP1B) 2151426) 0.10) 2.30E206) 0.09) 1.09E202)cg00251358) chr5) )) MIR3977) 2204472) 0.12) 2.00E204) 0.09) 2.67E201)cg23400056) chr5) )) NR3C1) 2141111) 0.11) 2.72E209) 0.07) 1.28E204)cg16168509) chr5) )) PCSK1) 82398) 0.12) 2.49E209) 0.13) 1.17E204)cg10189029) chr5) )) U6atac) 2237698) 0.13) 5.20E211) 0.06) 1.28E205)cg23400222) chr5) )) ZNF608) 2588206) 0.15) 3.07E206) 0.03) 1.39E202)227  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg24700316) chr6) BACH2) BACH2) 63065) 0.12) 5.33E208) 0.10) 8.72E204)cg07981495) chr6) CGA) CGA) 100) 0.15) 1.52E207) 0.08) 1.78E203)cg09261794) chr6) COL11A2) COL11A2) 22227) 0.12) 8.38E207) 0.03) 5.21E203)cg14692709) chr6) COL11A2) COL11A2) 8412) 0.11) 2.42E204) 0.02) 2.99E201)cg17295389) chr6) DDO) C6orf186) 14669) 0.11) 1.37E206) 0.10) 6.81E203)cg20422212) chr6) EYA4) EYA4) 13578) 0.10) 1.52E205) 0.07) 4.99E202)cg26501049) chr6) FAM184A) FAM184A) 2819) 0.12) 1.10E206) 0.09) 6.16E203)cg23978358) chr6) GFRAL) GFRAL) 2217) 0.11) 6.00E204) 0.04) 4.67E201)cg19722391) chr6) HIST1H1T) HIST1H1T) 2886) 0.14) 8.00E207) 0.08) 5.08E203)cg08468741) chr6) HLA2DMA) HLA2DMA) 3478) 0.10) 3.40E206) 0.04) 1.49E202)cg11377875) chr6) HMGCLL1) HMGCLL1) 2308) 0.12) 2.00E208) 0.05) 4.33E204)cg17066452) chr6) HSD17B8) HSD17B8) 1083) 0.12) 8.06E213) 0.21) 1.87E206)cg00415333) chr6) IL22RA2) IL22RA2) 1768) 0.15) 1.80E210) 0.04) 2.10E205)cg10791580) chr6)LOC285733) LOC100507203) 2674) 0.11) 3.33E205) 0.07) 8.60E202)cg21775668) chr6)LOC285733) LOC100507203) 21328) 0.11) 2.70E206) 0.15) 1.28E202)cg27423027) chr6)LOC285733) LOC100507203) 2744) 0.12) 2.56E206) 0.08) 1.18E202)cg08146841) chr6) NKAIN2) NKAIN2) 2483) 0.13) 3.99E207) 0.12) 3.30E203)cg26931063) chr6) OPN5) OPN5) 166) 0.12) 1.36E205) 0.03) 4.47E202)cg13401235) chr6) PHACTR1) LOC100130357) 120377) 0.11) 5.89E206) 0.08) 2.35E202)cg16734734) chr6) POPDC3) POPDC3) 2179) 0.11) 2.83E204) 0.04) 3.23E201)cg27541541) chr6) PRL) PRL) 2416) 0.11) 3.86E206) 0.08) 1.61E202)cg22328837) chr6) RUNX2) RUNX2) 32207) 0.11) 2.01E206) 0.06) 9.96E203)228  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg23963071) chr6) SERPINB9) SERPINB9) 2802) 0.12) 6.54E212) 0.07) 4.57E206)cg08278435) chr6) SLC17A2) SLC17A2) 1790) 0.11) 5.66E206) 0.11) 2.28E202)cg14224444) chr6) SLC35F1) BRD7P3) 2235345) 0.12) 2.72E209) 0.08) 1.29E204)cg13735018) chr6) SYNJ2) SYNJ2) 1234) 0.12) 4.85E207) 0.10) 3.65E203)cg10893014) chr6) TEAD3) TEAD3) 3625) 0.24) 3.02E208) 0.05) 6.06E204)cg14993900) chr6) TFAP2A) TFAP2A) 21272) 0.11) 4.59E209) 0.06) 1.81E204)cg19143211) chr6) TRDN) AL832096) 78972) 0.13) 8.53E205) 0.06) 1.84E201)cg25960769) chr6) TREML2) TREML2) 2886) 0.11) 4.23E206) 0.08) 1.74E202)cg12612406) chr6) TRIM40) TRIM40) 2425) 0.12) 4.02E205) 0.06) 1.04E201)cg11049439) chr6) ZBTB12) C2) 251) 0.10) 9.33E210) 0.04) 6.36E205)cg05590265) chr6) )) AK093513) 260584) 0.10) 1.51E206) 0.09) 7.81E203)cg04358957) chr6) )) C6orf195) 81243) 0.11) 5.96E205) 0.00) 1.45E201)cg01207473) chr6) )) C6orf99) 50322) 0.10) 4.46E205) 0.10) 1.09E201)cg16423738) chr6) )) IMPG1) 2611826) 0.11) 3.00E208) 0.13) 6.02E204)cg02479177) chr6) )) MARCKS) 2633700) 0.12) 7.98E205) 0.01) 1.74E201)cg16090790) chr6) )) MARCKS) 2503659) 0.15) 1.22E209) 0.11) 7.87E205)cg21950107) chr6) )) PIM1) 228399) 0.11) 1.42E206) 0.06) 7.05E203)cg08298351) chr6) )) SIM1) 24111) 20.11) 2.95E204) 20.02) 3.36E201)cg24717325) chr6) )) TRNA_Ala) 574) 0.10) 5.39E206) 0.05) 2.19E202)cg21545548) chr6) )) TRNA_Arg) 23329) 0.14) 4.05E210) 0.00) 3.76E205)cg14384014) chr6) )) U6) 227546) 0.13) 3.63E206) 0.13) 1.55E202)cg15059474) chr7) BAIAP2L1) BAIAP2L1) 40243) 0.11) 1.84E207) 0.03) 2.03E203)cg07776698) chr7) C7orf58) WNT16) 261468) 0.14) 4.86E207) 0.05) 3.66E203)cg24024511) chr7) C7orf71) C7orf71) 21143) 20.10) 1.21E204) 20.01) 2.22E201)229  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg22828602) chr7)CADPS2;RNF133) RNF133) 629) 0.11) 2.44E206) 0.05) 1.10E202)cg04590119) chr7) CHRM2) CHRM2) 9367) 0.11) 7.37E205) 0.16) 1.65E201)cg07116919) chr7) CHRM2) CHRM2) 4358) 0.13) 2.74E205) 0.04) 7.26E202)cg00832924) chr7) CTTNBP2) CTTNBP2) 64501) 0.13) 4.52E205) 0.01) 1.11E201)cg02132909) chr7) DYNC1I1) DYNC1I1) 21020) 0.10) 3.69E206) 0.02) 1.55E202)cg04156940) chr7) EGFR) EGFR) 20165) 0.11) 1.80E207) 0.04) 1.94E203)cg26311610) chr7) EIF4H) EIF4H) 4433) 0.10) 2.02E204) 0.08) 2.69E201)cg10336671) chr7) GHRHR) GHRHR) 2341) 0.12) 1.70E206) 0.07) 9.08E203)cg07834314) chr7) GNA12) AMZ1) 29046) 0.12) 1.63E208) 0.03) 3.67E204)cg08799766) chr7) GNA12) GNA12) 53061) 0.15) 7.17E208) 0.07) 1.14E203)cg16182691) chr7) GNAI1) GNAI1) 21183) 0.11) 6.74E208) 0.12) 1.07E203)cg01581050) chr7) HIPK2) CLEC2L) 47451) 0.11) 1.29E209) 0.21) 8.45E205)cg11355603) chr7) HIPK2) CLEC2L) 47574) 0.14) 3.17E205) 0.02) 8.41E202)cg11079619) chr7)INHBA;LOC285954) INHBA) 76) 0.21) 1.22E214) 0.05) 2.96E207)cg12979992) chr7) INSIG1) INSIG1) 11074) 0.16) 1.39E208) 0.08) 3.46E204)cg26800802) chr7) JHDM1D) JHDM1D) 17277) 0.13) 4.64E209) 0.09) 1.82E204)cg19323059) chr7)LOC285954) INHBA) 26638) 0.11) 3.40E207) 0.03) 2.80E203)cg18413237) chr7)LOC285954;INHBA) INHBA) 289) 0.12) 4.92E208) 0.09) 8.05E204)cg11370011) chr7) MIR29A) MIR29A) 58) 0.10) 1.03E209) 0.04) 7.04E205)cg07012178) chr7) PRKAG2) PRKAG2) 212171) 0.12) 2.69E214) 0.02) 2.97E207)cg22799757) chr7) SCRN1) SCRN1) 22666) 0.11) 3.68E208) 0.09) 6.63E204)230  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg06621691) chr7) SDK1) SDK1) 6498) 0.10) 5.93E206) 0.09) 2.40E202)cg25783987) chr7) SDK1) SDK1) 11407) 0.10) 3.93E205) 0.12) 9.74E202)cg25143609) chr7) STEAP4) STEAP4) 2673) 0.12) 4.29E210) 0.07) 3.82E205)cg23903035) chr7) STRA8) STRA8) 15719) 0.10) 7.08E204) 0.03) 4.95E201)cg06480942) chr7) STX1A) MIR4284) 29617) 0.14) 8.74E208) 0.04) 1.29E203)cg03764092) chr7) TBXAS1) PARP12) 72668) 0.15) 1.83E208) 0.03) 4.14E204)cg24431161) chr7) TBXAS1) TBXAS1) 6305) 0.17) 2.02E208) 0.09) 4.46E204)cg09507697) chr7) TSGA14) CEP41) 3002) 0.11) 8.87E208) 0.09) 1.34E203)cg18783886) chr7) ZNF783) ZNF783) 5277) 0.11) 2.22E208) 0.01) 4.64E204)cg01116873) chr7) )) 7SK) 237142) 0.11) 1.85E204) 0.05) 2.60E201)cg14605117) chr7) )) AMZ1) 24105) 0.18) 2.87E212) 0.12) 3.54E206)cg16568084) chr7) )) BET1) 2177873) 0.15) 1.55E210) 0.08) 1.92E205)cg03188976) chr7) )) CALD1) 220299) 0.12) 7.90E211) 0.05) 1.64E205)cg11644973) chr7) )) DNAJB6) 243730) 0.11) 7.35E206) 0.02) 2.65E202)cg22497133) chr7) )) DQ599872) 2668995) 0.12) 2.74E204) 0.05) 3.19E201)cg23516370) chr7) )) DQ599872) 2727456) 0.12) 5.82E205) 0.12) 1.43E201)cg00159987) chr7) )) INHBA) 2106033) 0.10) 6.36E206) 0.08) 2.42E202)cg18285819) chr7) )) INMT) 2970) 0.11) 5.04E207) 0.09) 3.72E203)cg02813710) chr7) )) SRPK2) 239865) 0.11) 4.01E205) 0.04) 1.03E201)cg18018990) chr7) )) U6) 1676) 0.10) 1.14E205) 0.11) 3.68E202)cg12996622) chr7) )) X06774) 2430) 0.11) 1.22E208) 0.11) 2.97E204)cg00659129) chr8) ASAP1) ASAP1) 41292) 0.10) 1.08E209) 0.07) 7.27E205)cg19366147) chr8) CLN8) CLN8) 5791) 0.10) 2.45E208) 0.07) 5.36E204)cg14854503) chr8)CSGALNACT1) CSGALNACT1) 211) 0.19) 1.45E209) 0.06) 9.01E205)231  Probe& CHR&Gene&Name&Closest&TSS&Gene&Name&Distance&closest&TSS&EOPE&Δβ&Discovery&Cohort&EOPE&vs&Preterm&p?value&Discovery&Cohort&EOPE&Δβ&&Validation&Cohort&EOPE&vs&Preterm&p?value&Validation&Cohort&cg13175060) chr8) ERICH1) ERICH1) 12519) 0.15) 4.81E209) 0.00) 1.87E204)cg26366616) chr8) PDLIM2) PDLIM2) 1845) 0.13) 1.02E210) 0.01) 1.74E205)cg27271298) chr8) RIMS2) RIMS2) 166826) 0.12) 1.08E204) 0.02) 2.10E201)cg04804543) chr8) SLC45A4) SLC45A4) 5246) 0.10) 1.75E206) 0.12) 9.10E203)cg00592695) chr8) ST3GAL1) ST3GAL1) 38498) 0.28) 4.51E214) 0.12) 5.67E207)cg00736681) chr8) ST3GAL1) ST3GAL1) 38131) 0.16) 7.38E212) 0.04) 5.26E206)cg01569173) chr8) TSPYL5) TSPYL5) 1266) 0.13) 1.05E204) 0.05) 2.07E201)cg04777352) chr8) )) 5S_rRNA) 60513) 0.10) 3.17E207) 0.07) 2.60E203)cg19883388) chr8) )) AZIN1) 233533) 0.17) 1.14E208) 0.03) 2.89E204)cg01874183) chr8) )) ENPP2) 48037) 0.12) 6.09E207) 0.03) 4.31E203)cg17661601) chr8) )) FDFT1) 23497) 0.10) 1.58E206) 0.09) 8.29E203)cg11996860) chr8) )) FLJ42969) 27534) 0.10) 1.94E208) 0.04) 4.24E204)cg14556425) chr8) )) IL7) 23353) 0.16) 2.55E205) 0.06) 6.85E202)cg26509870) chr8) )) PHYHIP) 13685) 0.19) 3.03E212) 0.02) 3.54E206)cg01713086) chr8) )) ZNF395) 224436) 0.14) 2.98E212) 0.00) 3.54E206)cg13562353) chr9) CCL27) CCL27) 284) 0.12) 4.60E209) 0.01) 1.81E204)cg14350701) chr9) NACC2) NACC2) 17419) 0.13) 1.16E206) 0.06) 6.20E203)cg00992055) chr9) OLFML2A) OLFML2A) 8446) 0.10) 1.08E209) 0.11) 7.29E205)cg14334810) chr9) PLIN2) PLIN2) 1310) 0.13) 3.31E211) 0.12) 1.09E205)cg13715502) chr9) PTPN3) PTPN3) 1817) 0.10) 1.25E208) 0.13) 3.17E204)cg25989745) chr9) TYRP1) TYRP1) 259) 0.11) 6.75E205) 0.11) 1.56E201)cg02169542) chr9) )) SLC24A2) 2143801) 0.12) 2.90E210) 0.02) 2.76E205)232   Supplementary Figure D.5. A) The concordance rate (sites where DNAm change is in the same direction in both cohorts) for different FDR and ∆β thresholds. Concordance rates (%) are plotted on the y-axis and FDR thresholds are on the x-axis. B) The number of identified hits for different FDR and ∆β thresholds. The number of hits is plotted on the y-axis and the FDR thresholds are on the x-axis.233   Supplementary Table D.2. Site and gene information on the 244 CpG sites that were differentially methylated between methylation cluster 2 and methylation cluster 3 in both the discovery and validation cohorts.  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg00251358) chr5) MIR3977) 2204472) 0.08) 1.36E202) 0.05) 2.83E202)cg00445523) chr3) CACNA2D2) 219564) 0.10) 1.14E202) 0.03) 2.31E202)cg00659129) chr8) ASAP1) 41292) 0.07) 7.98E203) 0.05) 5.82E205)cg00736681) chr8) ST3GAL1) 38131) 0.05) 4.03E202) 0.10) 1.54E210)cg00762397) chr1) PRELP) 9128) 0.05) 6.95E203) 0.08) 5.00E210)cg00964446) chr1) PPFIA4) 22274) 0.10) 1.89E202) 0.06) 1.68E206)cg01127412) chr3) XIRP1) 1228) 0.04) 3.28E202) 0.04) 7.39E204)cg01334432) chr1) MIR205) 2894) 0.05) 4.40E202) 0.02) 2.53E202)cg01412654) chr3) PPARG) 21264) 0.06) 8.35E203) 0.04) 2.98E203)cg01427300) chr13) CAB39L) 53856) 0.05) 3.98E202) 0.04) 5.42E204)cg01491071) chr2) BC047484) 2160539) 0.09) 2.02E202) 0.10) 1.82E208)cg01571842) chr3) SLMAP) 52636) 0.08) 1.64E202) 0.04) 1.74E202)cg01874183) chr8) ENPP2) 48037) 0.08) 2.61E202) 0.08) 3.45E206)cg02023717) chr17) ERBB2) 2305) 0.09) 1.97E203) 0.05) 1.18E205)cg02058483) chr11) SLC1A2) 21308) 0.04) 3.89E202) 0.05) 3.10E203)cg02132909) chr7) DYNC1I1) 21020) 0.07) 3.30E202) 0.05) 2.66E203)cg02169542) chr9) SLC24A2) 2143801) 0.06) 1.18E202) 0.06) 1.30E204)cg02287004) chr2) RDH14) 135479) 0.10) 1.07E204) 0.06) 3.44E205)234  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg02479177) chr6) MARCKS) 2633700) 0.08) 2.33E203) 0.09) 2.52E206)cg02843417) chr16) IRF8) 13462) 0.07) 2.48E204) 0.05) 1.42E205)cg03174768) chr4) LIMCH1) 7106) 0.04) 1.40E202) 0.03) 2.86E202)cg03180819) chr2) SCN3A) 21022) 0.12) 2.12E203) 0.06) 2.61E203)cg03349057) chr3) ITIH4) 5151) 0.11) 5.24E203) 0.04) 2.27E203)cg03511671) chr1) PPFIA4) 2790) 0.08) 2.92E202) 0.08) 6.87E206)cg03539735) chr14) EVL) 235492) 0.05) 3.80E203) 0.06) 9.22E209)cg03650068) chr1) RHD) 22489) 0.07) 2.68E202) 0.04) 3.40E203)cg03764092) chr7) PARP12) 72668) 0.10) 4.78E202) 0.07) 4.69E206)cg03851970) chr3) MITF) 6562) 0.18) 3.06E204) 0.06) 1.16E202)cg03865281) chr10) LIPJ) 18153) 0.12) 9.88E204) 0.07) 3.11E204)cg03932039) chr10) LHPP) 2402) 0.08) 1.49E202) 0.06) 1.24E206)cg04038163) chr4) CXCL9) 3486) 0.09) 9.39E203) 0.07) 7.98E207)cg04156940) chr7) EGFR) 20165) 0.05) 2.08E202) 0.07) 5.71E208)cg04543510) chr5) AGXT2) 22416) 0.06) 2.19E203) 0.05) 1.89E204)cg04662250) chr17) C17orf103) 2921) 0.06) 3.36E203) 0.05) 3.11E205)cg04708790) chr12) OAS1) 10937) 0.07) 1.94E203) 0.04) 4.70E202)cg04804772) chr10) CALHM3) 124) 0.05) 1.44E202) 0.04) 5.34E205)cg04897892) chr16) CMIP) 35361) 0.08) 4.62E202) 0.06) 3.85E204)cg04945434) chr11) SnoMBII_202) 2165105) 0.09) 1.58E203) 0.09) 2.65E207)cg04981492) chr19) SYDE1) 500) 0.04) 4.80E202) 0.03) 4.67E202)cg05046721) chr1) KAZN) 242369) 0.11) 2.64E203) 0.08) 2.58E205)cg05074732) chr2) STK39) 292019) 0.09) 3.88E205) 0.05) 1.01E203)235  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg05211068) chr5) AK091866) 35078) 0.04) 3.02E202) 0.05) 1.06E203)cg05371908) chr1) CACHD1) 236888) 0.12) 4.50E203) 0.06) 4.77E203)cg05394210) chr11) DCDC5) 21981) 0.15) 9.54E204) 0.04) 4.91E202)cg05507514) chr3) ARHGEF26) 3698) 0.07) 2.62E204) 0.06) 1.50E204)cg05590265) chr6) AK093513) 260584) 0.07) 3.01E203) 0.07) 2.35E204)cg05663031) chr3) BCL6) 564) 0.05) 2.19E202) 0.07) 3.84E206)cg05819416) chr1) PROX12AS1) 2890) 0.12) 1.86E204) 0.06) 1.07E203)cg06220112) chr4) LOC100505875) 111386) 0.15) 8.05E204) 0.06) 1.72E203)cg06375062) chr2) LOC100128590) 4485) 0.10) 2.30E204) 0.03) 7.96E203)cg06495803) chr12) CD27) 2561) 0.06) 1.42E202) 0.06) 3.54E206)cg06685875) chr5) PCSK1) 4668) 0.06) 3.29E203) 0.07) 5.23E206)cg06691616) chr4) SLC7A11) 2367595) 0.12) 1.08E203) 0.05) 2.11E203)cg06912936) chr5) KCNIP1) 4150) 0.06) 3.34E202) 0.06) 2.30E208)cg07012178) chr7) PRKAG2) 212171) 0.04) 1.60E202) 0.04) 2.31E204)cg07116919) chr7) CHRM2) 4358) 0.09) 7.61E203) 0.09) 3.80E206)cg07168214) chr17) ZBTB4) 2832) 0.04) 2.83E202) 0.05) 9.79E206)cg07177437) chr1) ITLN1) 4583) 0.08) 4.76E202) 0.07) 6.48E206)cg07233933) chr5) CPEB4) 219269) 0.04) 5.61E203) 0.05) 5.74E206)cg07344172) chr1) KIF21B) 25666) 0.10) 1.29E203) 0.06) 2.07E205)cg07530172) chr11) FAM76B) 91581) 0.09) 1.10E203) 0.06) 1.40E204)cg07579831) chr11) SHANK2) 248682) 0.12) 2.02E203) 0.06) 2.42E203)cg07762788) chr16) TMCO7) 244202) 0.06) 2.25E202) 0.05) 1.84E203)cg07776698) chr7) WNT16) 261468) 0.13) 2.46E204) 0.07) 7.14E205)236  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg07834314) chr7) AMZ1) 29046) 0.07) 2.70E203) 0.07) 1.56E208)cg07913153) chr1) LGALS8) 2752) 0.05) 2.42E202) 0.02) 4.46E202)cg07989490) chr10) GATA3) 20360) 0.05) 7.40E203) 0.04) 2.30E203)cg08116915) chr4) CPZ) 2662) 0.07) 8.93E203) 0.06) 7.04E206)cg08146841) chr6) NKAIN2) 2483) 0.11) 6.00E204) 0.05) 2.83E202)cg08250187) chr2) 7SK) 2178048) 0.09) 4.06E203) 0.06) 3.41E205)cg08278435) chr6) SLC17A2) 1790) 0.08) 1.46E202) 0.07) 3.46E205)cg08799766) chr7) GNA12) 53061) 0.07) 2.84E202) 0.04) 1.25E202)cg08872493) chr1) HTR1D) 2195) 0.05) 3.74E202) 0.05) 5.99E203)cg08885854) chr1) 7SK) 255636) 0.18) 2.91E205) 0.06) 6.60E204)cg08991927) chr5) PPP2R2B) 21105) 0.14) 7.53E204) 0.05) 7.42E203)cg09038266) chr16) MAF) 478664) 0.10) 7.74E203) 0.06) 4.38E203)cg09261794) chr6) COL11A2) 22227) 0.10) 1.46E202) 0.08) 6.20E206)cg09367174) chr2) NEB) 2165) 0.16) 5.11E205) 0.05) 2.19E203)cg09476306) chr22) YPEL1) 3736) 0.12) 1.35E203) 0.05) 1.71E204)cg09500437) chr2) MIR4265) 255132) 0.07) 4.97E202) 0.07) 1.20E206)cg09623279) chr10) TMEM72) 211029) 0.10) 2.52E203) 0.06) 1.14E205)cg09624551) chr3) AK097190) 21192) 0.09) 4.01E203) 0.05) 2.59E204)cg09638264) chr2) WDR69) 2604) 20.07) 6.59E203) 0.06) 1.22E203)cg09648454) chr16) ABCC12) 22825) 0.06) 4.93E202) 0.05) 1.26E204)cg09664054) chr1) HIVEP3) 2128108) 0.12) 7.80E203) 0.09) 4.66E207)cg09781307) chr11) EF565102) 7804) 0.10) 2.56E204) 0.06) 1.71E203)cg10092779) chr1) BC048114) 21492) 0.07) 1.98E202) 0.06) 3.62E206)237  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg10144104) chr3) CACNA2D2) 21996) 0.11) 2.85E202) 0.08) 1.00E206)cg10246581) chr16) CMIP) 7218) 0.06) 1.00E202) 0.05) 2.26E203)cg10322735) chr4) LINC00290) 2349) 0.11) 3.44E202) 0.06) 1.01E202)cg10336671) chr7) GHRHR) 2341) 0.14) 9.67E204) 0.05) 1.14E203)cg10440877) chr2) CREB1) 216140) 0.04) 4.16E202) 0.04) 9.13E205)cg10455413) chr2) AGAP1) 49872) 0.05) 3.21E202) 0.05) 5.85E207)cg10472840) chr5) ESM1) 75353) 0.09) 2.41E204) 0.04) 3.83E202)cg10620457) chr1) C8B) 2422) 0.10) 3.40E203) 0.05) 8.60E203)cg10791580) chr6) LOC100507203) 2674) 0.09) 1.34E204) 0.06) 4.04E204)cg11348701) chr2) GAD1) 21353) 0.09) 2.33E203) 0.05) 4.47E204)cg11370011) chr7) MIR29A) 58) 0.06) 5.59E203) 0.07) 1.28E205)cg11377875) chr6) HMGCLL1) 2308) 0.07) 7.15E203) 0.05) 1.28E202)cg11451315) chr10) ANKRD30A) 2436731) 0.08) 9.00E203) 0.05) 2.98E203)cg11693581) chr10) KIAA1598) 2509) 0.05) 2.66E202) 0.03) 3.98E202)cg12140749) chr14) C14orf181) 21028) 0.04) 2.48E202) 0.05) 9.08E206)cg12156272) chr14) C14orf184) 2873) 0.12) 7.12E203) 0.03) 3.60E202)cg12290217) chr15) AL109706) 28742) 0.05) 1.72E202) 0.06) 1.78E205)cg12387464) chr11) OR8D4) 21203) 0.19) 2.55E206) 0.06) 1.99E203)cg12493761) chr12) PRICKLE1) 2269685) 0.07) 5.57E203) 0.05) 2.49E203)cg12686754) chr2) MIR216A) 2389) 0.09) 7.21E203) 0.06) 4.14E205)cg12732548) chr15) C15orf52) 21239) 0.05) 1.78E202) 0.07) 9.91E208)cg12750431) chr10) DKFZp547J0510) 22817) 0.12) 5.39E203) 0.07) 1.39E204)cg12792952) chr1) CACHD1) 279088) 0.16) 3.35E203) 0.06) 4.09E203)238  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg12833765) chr2) MIR3126) 140065) 0.08) 6.64E203) 0.07) 2.32E205)cg12845923) chr4) PDGFRA) 23469) 0.09) 2.39E203) 0.05) 7.62E203)cg12961926) chr17) TBCD) 214400) 0.07) 7.60E203) 0.04) 6.09E204)cg13056990) chr1) AJAP1) 20995) 0.05) 3.99E202) 0.06) 5.58E204)cg13107839) chr1) KAZN) 277501) 0.13) 1.08E202) 0.06) 1.95E202)cg13196578) chr1) PLEKHM2) 22659) 0.06) 1.78E202) 0.05) 6.70E205)cg13301676) chr5) MCC) 124452) 0.14) 4.12E203) 0.05) 3.39E202)cg13345122) chr2) B3GALT1) 187336) 0.11) 3.64E203) 0.04) 2.17E202)cg13401235) chr6) LOC100130357) 120377) 0.10) 6.36E203) 0.06) 1.36E202)cg13546308) chr3) C3orf79) 163872) 0.11) 5.65E203) 0.06) 1.12E204)cg13735018) chr6) SYNJ2) 1234) 0.06) 3.92E202) 0.06) 2.39E207)cg13785473) chr4) MFSD10) 5187) 0.06) 2.19E203) 0.07) 4.39E205)cg13897222) chr1) C1orf140) 59852) 0.11) 2.57E202) 0.06) 3.28E203)cg14224444) chr6) BRD7P3) 2235345) 0.11) 3.24E203) 0.07) 3.87E205)cg14500854) chr10) PRKG1) 252335) 0.08) 1.77E202) 0.05) 6.18E203)cg14556425) chr8) IL7) 23353) 0.20) 2.73E204) 0.09) 5.15E204)cg14692709) chr6) COL11A2) 8412) 0.10) 4.80E203) 0.08) 1.02E203)cg14851263) chr15) MYO5A) 21335) 0.09) 2.68E202) 0.05) 1.27E202)cg14938029) chr20) CSTL1) 967) 0.09) 1.03E202) 0.05) 7.46E204)cg15005526) chr2) SMYD5) 23223) 0.09) 9.26E203) 0.07) 1.59E206)cg15059474) chr7) BAIAP2L1) 40243) 0.06) 6.05E203) 0.04) 8.44E203)cg15282973) chr12) SCARB1) 2266) 0.06) 1.75E202) 0.07) 1.14E205)cg15410418) chr5) PDE4D) 2794) 0.10) 2.39E204) 0.06) 7.94E204)239  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg15432510) chr17) AK127919) 5343) 0.12) 6.44E204) 0.09) 6.46E206)cg15433043) chr11) MYO7A) 21400) 0.06) 3.61E203) 0.05) 1.78E204)cg15492834) chr1) C1orf226) 2462) 0.08) 3.88E203) 0.06) 5.98E207)cg15534461) chr12) PEBP1) 1164) 0.07) 1.82E203) 0.03) 4.03E202)cg16056164) chr1) PCNXL2) 217308) 0.10) 9.75E203) 0.04) 1.17E202)cg16168509) chr5) PCSK1) 82398) 0.11) 1.06E203) 0.05) 4.43E205)cg16177314) chr4) GYPB) 21113) 0.10) 1.66E204) 0.04) 1.22E202)cg16178353) chr5) ATP6AP1L) 29514) 0.13) 1.20E203) 0.04) 2.51E202)cg16243351) chr11) SLC1A2) 211980) 0.09) 4.21E203) 0.05) 8.06E205)cg16253629) chr4) ANK2) 36978) 0.07) 2.46E203) 0.08) 4.40E206)cg16291750) chr3) BC035826) 2502) 0.09) 1.98E204) 0.04) 3.64E203)cg16326590) chr2) VWA3B) 1264) 0.04) 2.99E202) 0.07) 1.10E208)cg16353318) chr16) CMIP) 35208) 0.06) 4.27E202) 0.04) 1.40E204)cg16423738) chr6) IMPG1) 2611826) 0.09) 1.14E203) 0.07) 1.61E206)cg16468174) chr16) BC013599) 4809) 0.06) 1.16E204) 0.08) 6.07E209)cg16565154) chr1) TNFAIP8L22SCNM1) 242) 0.05) 2.21E202) 0.03) 4.39E202)cg16926302) chr1) ST6GALNAC5) 77721) 0.12) 8.18E204) 0.06) 1.77E202)cg16964478) chr12) BC041342) 21228) 0.12) 7.80E204) 0.06) 5.97E203)cg17066452) chr6) HSD17B8) 1083) 0.05) 2.39E202) 0.05) 1.01E206)cg17108958) chr1) HLX) 76803) 0.08) 1.03E203) 0.06) 2.17E203)cg17284609) chr11) SOX6) 2260) 0.08) 5.44E204) 0.05) 2.01E204)cg17295389) chr6) C6orf186) 14669) 0.09) 1.03E203) 0.04) 3.07E203)cg17357548) chr15) ATP8B4) 218410) 0.07) 6.48E203) 0.04) 9.09E203)240  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg17394304) chr3) BCL6) 312) 0.05) 2.00E202) 0.06) 4.39E206)cg17425351) chr11) 5S_rRNA) 67887) 0.10) 1.37E203) 0.08) 7.88E205)cg18018990) chr7) U6) 1676) 0.13) 7.65E204) 0.03) 3.66E202)cg18190824) chr5) FAM105B) 56244) 20.03) 3.51E202) 0.10) 4.08E211)cg18285819) chr7) INMT) 2970) 0.08) 1.64E202) 0.07) 2.03E209)cg18512205) chr15) GPR176) 29534) 0.09) 1.92E205) 0.04) 5.28E203)cg18689240) chr12) KRT7) 22214) 0.06) 3.53E202) 0.08) 8.30E205)cg18813778) chr3) KCNH8) 273686) 0.12) 9.77E203) 0.05) 1.36E202)cg19024161) chr10) PBLD) 11830) 0.10) 1.74E203) 0.04) 3.59E203)cg19118951) chr21) LINC00310) 22093) 0.04) 4.22E202) 0.05) 5.10E203)cg19183252) chr4) BC038750) 59729) 0.10) 7.52E203) 0.04) 3.72E202)cg19192065) chr10) LOC100188947) 36126) 0.09) 1.97E203) 0.06) 3.83E205)cg19291696) chr13) USP12) 223404) 0.06) 4.23E202) 0.04) 3.11E202)cg19366147) chr8) CLN8) 5791) 0.03) 1.88E202) 0.04) 7.37E203)cg19674091) chr15) IDH2) 161) 0.10) 3.89E202) 0.05) 1.36E202)cg19786983) chr15) AK094929) 68936) 0.09) 3.63E203) 0.07) 1.46E206)cg19883388) chr8) AZIN1) 233533) 0.11) 3.24E203) 0.08) 1.24E204)cg20054786) chr4) TLR1) 1046) 0.08) 1.93E202) 0.04) 1.49E203)cg20080983) chr22) C22orf31) 27636) 0.05) 1.61E202) 0.05) 2.15E204)cg20090782) chr18) SETBP1) 62854) 0.06) 1.01E202) 0.05) 2.36E206)cg20092728) chr11) SLC5A12) 507) 0.09) 6.44E203) 0.06) 3.76E203)cg20389635) chr12) PTHLH) 2840) 0.12) 1.53E203) 0.06) 5.56E203)cg20395789) chr11) SLN) 7775) 0.05) 4.11E203) 0.06) 1.42E205)241  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg20438445) chr17) RAB5C) 22295) 0.11) 1.09E203) 0.03) 1.60E202)cg20497304) chr2) ATP5G3) 248827) 0.10) 1.18E204) 0.08) 8.67E207)cg20570458) chr3) KALRN) 233738) 0.05) 4.97E202) 0.06) 4.55E204)cg20733250) chr11) GRIK4) 25385) 0.11) 3.24E203) 0.08) 1.59E204)cg20782816) chr3) CAMK1) 3608) 0.05) 3.70E202) 0.08) 9.09E210)cg20822693) chr2) SLC16A14) 2684) 0.07) 8.74E203) 0.06) 7.69E206)cg20991819) chr1) U3) 97472) 0.09) 1.87E204) 0.07) 5.37E209)cg21545548) chr6) TRNA_Arg) 23329) 0.03) 2.25E202) 0.02) 1.25E202)cg21733367) chr15) DQ575076) 2428) 0.08) 2.33E202) 0.07) 1.22E203)cg21760128) chr3) LARS2) 22471) 0.06) 3.96E202) 0.04) 1.76E202)cg21775668) chr6) LOC100507203) 21328) 0.09) 5.56E205) 0.03) 6.76E203)cg21823502) chr17) NPB) 23641) 0.04) 3.19E202) 0.07) 5.21E206)cg21837826) chr1) FAM163A) 4434) 0.09) 1.07E203) 0.05) 1.61E202)cg21977522) chr3) U6atac) 20133) 0.16) 1.40E205) 0.07) 7.76E204)cg22234930) chr15) PKM2) 1278) 0.06) 1.10E202) 0.04) 2.32E202)cg22245446) chr2) DIRC3) 149101) 0.10) 9.34E204) 0.08) 3.84E206)cg22497133) chr7) DQ599872) 2668995) 0.22) 6.30E204) 0.06) 2.24E202)cg22520372) chr4) NPY5R) 67) 0.08) 6.01E203) 0.07) 8.01E204)cg22671342) chr1) IL24) 6552) 0.09) 2.81E204) 0.04) 4.03E203)cg22799757) chr7) SCRN1) 22666) 0.09) 7.83E204) 0.05) 2.43E205)cg22828602) chr7) RNF133) 629) 0.09) 7.89E204) 0.05) 1.55E202)cg22860917) chr2) GALM) 26300) 0.05) 5.76E203) 0.07) 5.58E205)cg22981158) chr19) LOC100506068) 27219) 0.12) 1.40E202) 0.10) 4.45E207)242  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg23101885) chr2) RAD51AP2) 3681) 0.11) 2.48E205) 0.05) 5.10E205)cg23104436) chr11) SnoMBII_202) 2335308) 0.08) 4.39E203) 0.07) 9.56E207)cg23128495) chr12) NUAK1) 279328) 0.09) 1.20E202) 0.08) 5.42E206)cg23255934) chr2) TRNA_Pseudo) 2194088) 0.11) 4.73E202) 0.05) 4.57E202)cg23309843) chr5) CTXN3) 2881) 0.13) 4.69E203) 0.07) 5.33E206)cg23330212) chr11) BDNF) 8499) 0.11) 5.38E204) 0.04) 3.01E203)cg23697796) chr10) LOC100188947) 36243) 0.08) 1.19E202) 0.06) 1.42E203)cg23831864) chr4) U6) 2282236) 0.15) 2.20E204) 0.04) 1.70E202)cg23963071) chr6) SERPINB9) 2802) 0.06) 9.92E203) 0.06) 1.30E204)cg24233594) chr10) AX747408) 243064) 0.07) 7.61E203) 0.05) 8.03E204)cg24340655) chr1) RGS4) 282770) 0.08) 4.99E203) 0.07) 1.08E204)cg24700316) chr6) BACH2) 63065) 0.09) 8.69E204) 0.07) 7.55E207)cg24715412) chr4) STOX2) 27739) 0.11) 2.58E202) 0.04) 1.05E202)cg24761668) chr2) SULT1C4) 31701) 0.09) 2.62E203) 0.05) 2.18E202)cg24938761) chr19) MOB3A) 4863) 0.05) 2.07E202) 0.04) 1.10E202)cg24995347) chr2) SP110) 803) 0.10) 3.70E204) 0.04) 2.35E203)cg25103160) chr12) SNORA70) 2123739) 0.05) 1.06E202) 0.07) 6.70E210)cg25103772) chr1) PAPPA2) 81399) 0.09) 1.35E202) 0.04) 2.37E202)cg25212701) chr1) C1orf226) 145) 0.08) 2.37E202) 0.07) 1.36E206)cg25298189) chr19) ARID3A) 9223) 20.07) 2.67E203) 0.03) 6.51E203)cg25390063) chr12) AK128707) 233405) 0.11) 6.98E204) 0.07) 2.99E205)cg25429446) chr4) ANK2) 99807) 0.10) 1.19E204) 0.03) 2.30E202)cg25469314) chr1) SLC5A9) 2162) 0.13) 1.51E203) 0.09) 6.90E206)243  Probe& Chromosome& Closest&TSS&Gene&Name& Distance_closest_TSS&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&&p?value&Discovery&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&Δβ&Validation&Cohort&Methylation&Cluster&3&vs&Methylation&Cluster&2&p?value&Validation&Cohort&cg25960769) chr6) TREML2) 2886) 0.11) 1.57E202) 0.04) 2.11E202)cg25989745) chr9) TYRP1) 259) 0.07) 1.30E202) 0.07) 1.07E203)cg26020161) chr1) GSTM3) 2763) 0.08) 3.83E204) 0.06) 1.61E205)cg26276667) chr10) NUDT9P1) 23452) 0.07) 7.06E203) 0.08) 5.25E205)cg26331247) chr20) C20orf203) 612) 0.06) 4.95E202) 0.06) 2.17E204)cg26365254) chr1) KIRREL) 210944) 0.11) 3.87E203) 0.06) 1.18E203)cg26378984) chr11) Mir_652) 224329) 0.09) 4.12E203) 0.04) 9.95E204)cg26548682) chr1) AK123450) 22942) 0.04) 3.46E202) 0.06) 3.73E205)cg26548883) chr17) CCL142CCL15) 2815) 0.13) 2.75E204) 0.07) 1.68E203)cg26651514) chr13) STARD13) 24833) 20.06) 3.00E202) 0.06) 4.09E204)cg26851803) chr12) 5S_rRNA) 267346) 0.09) 1.41E203) 0.08) 1.84E207)cg26861706) chr15) CHSY1) 41736) 0.05) 6.75E203) 0.05) 4.45E208)cg27223882) chr2) RPRM) 89976) 0.12) 7.77E203) 0.05) 1.55E202)cg27271298) chr8) RIMS2) 166826) 0.12) 6.63E203) 0.09) 7.55E204)cg27321175) chr10) ITGA8) 72761) 0.10) 1.12E202) 0.04) 2.04E202)cg27423027) chr6) LOC100507203) 2744) 0.11) 3.82E205) 0.05) 1.22E203)cg27442676) chr4) SLC4A4) 17665) 0.12) 3.73E204) 0.04) 9.71E203)cg27492725) chr2) PDE1A) 221964) 0.12) 1.14E205) 0.06) 9.76E204)cg27502580) chr3) DD413694) 132454) 0.08) 6.93E203) 0.03) 4.33E202) 244  Supplementary Table D.3. Site and gene information on the 207 sites differentially methylated between EOPE methylation subcluster 1 and EOPE methylation subcluster 2 in both the discovery and validation cohorts.  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg00445523) chr3) 219564) CACNA2D2) 0.07) 4.90E204) 20.06) 4.74E202)cg00736681) chr8) 38131) ST3GAL1) 0.04) 1.21E202) 20.06) 1.02E202)cg00762397) chr1) 9128) PRELP) 0.03) 3.20E202) 20.09) 1.67E205)cg00832924) chr7) 64501) CTTNBP2) 0.10) 9.69E203) 20.12) 4.29E202)cg00964446) chr1) 22274) PPFIA4) 0.04) 3.14E202) 20.12) 9.61E203)cg01116873) chr7) 237142) 7SK) 0.08) 7.85E203) 20.20) 1.15E203)cg01140247) chr10) 217702) Metazoa_SRP) 0.04) 2.44E203) 20.08) 3.66E204)cg01238639) chr4) 381987) TRAM1L1) 0.06) 4.96E202) 20.22) 1.26E205)cg01413880) chr2) 28732) Mir_548) 0.06) 2.74E203) 20.05) 5.81E203)cg01491071) chr2) 2160539) BC047484) 0.06) 7.40E204) 20.13) 4.58E204)cg01571842) chr3) 52636) SLMAP) 0.10) 3.65E202) 20.09) 3.67E202)cg01874183) chr8) 48037) ENPP2) 0.04) 1.27E202) 20.12) 7.35E204)cg02132909) chr7) 21020) DYNC1I1) 0.05) 1.25E203) 20.09) 6.13E204)cg02169542) chr9) 2143801) SLC24A2) 0.09) 4.70E206) 20.11) 8.98E204)cg02287004) chr2) 135479) RDH14) 0.04) 3.78E202) 20.13) 7.54E206)cg02304751) chr4) 24133) TSPAN5) 0.08) 1.53E205) 20.11) 1.16E203)cg02448743) chr10) 1744) KAZALD1) 0.05) 7.11E203) 20.05) 4.08E202)cg02479177) chr6) 2633700) MARCKS) 0.03) 2.34E202) 20.12) 9.64E204)cg02625481) chr2) 64374) FRZB) 0.06) 3.39E203) 20.14) 4.44E205)cg03180819) chr2) 21022) SCN3A) 0.08) 3.20E203) 20.10) 2.34E202) 245  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg03349057) chr3) 5151) ITIH4) 0.06) 2.89E202) 20.09) 1.66E202)cg03511671) chr1) 2790) PPFIA4) 0.05) 4.19E203) 20.09) 1.36E203)cg03539735) chr14) 235492) EVL) 0.05) 1.27E204) 20.06) 3.15E202)cg03728457) chr15) 8696) CCDC33) 0.10) 6.44E205) 20.06) 5.58E203)cg03764092) chr7) 72668) PARP12) 0.12) 3.75E204) 20.15) 3.60E203)cg03851970) chr3) 6562) MITF) 0.09) 1.29E202) 20.21) 1.53E204)cg03865281) chr10) 18153) LIPJ) 0.06) 3.40E202) 20.13) 5.26E204)cg04038163) chr4) 3486) CXCL9) 0.06) 1.29E203) 20.09) 1.05E202)cg04103490) chr1) 21092) HIST3H3) 0.07) 1.88E204) 20.09) 9.53E203)cg04413904) chr16) 234276) SSTR5) 0.10) 2.09E202) 20.12) 2.09E202)cg04543510) chr5) 22416) AGXT2) 0.04) 2.25E202) 20.09) 2.66E204)cg04590119) chr7) 9367) CHRM2) 0.06) 1.33E202) 20.08) 7.50E203)cg04597985) chr4) 2203) BC031238) 0.10) 2.98E204) 20.20) 1.25E204)cg04777352) chr8) 60513) 5S_rRNA) 0.08) 2.22E204) 20.06) 8.61E204)cg04945434) chr11) 2165105) SnoMBII_202) 0.05) 1.03E203) 20.10) 2.75E204)cg05046721) chr1) 242369) KAZN) 0.05) 1.38E202) 20.10) 6.08E203)cg05074732) chr2) 292019) STK39) 0.07) 3.17E204) 20.14) 2.92E204)cg05199904) chr5) 2151426) MAP1B) 0.06) 6.40E204) 20.11) 1.08E205)cg05371908) chr1) 236888) CACHD1) 0.07) 3.60E203) 20.17) 1.32E205)cg05507514) chr3) 3698) ARHGEF26) 0.07) 1.04E204) 20.09) 1.70E204)cg05590265) chr6) 260584) AK093513) 0.09) 1.99E203) 20.12) 2.37E204)cg05819416) chr1) 2890) PROX12AS1) 0.07) 4.15E203) 20.16) 3.87E204)cg06220112) chr4) 111386) LOC100505875) 0.07) 8.73E203) 20.22) 6.67E205) 246  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg06340382) chr3) 288610) P2RY1) 0.08) 5.82E203) 20.11) 7.45E203)cg06375062) chr2) 4485) LOC100128590) 0.07) 1.07E203) 20.16) 1.38E205)cg06495803) chr12) 2561) CD27) 0.04) 3.03E202) 20.09) 9.78E206)cg06621691) chr7) 6498) SDK1) 0.07) 5.30E203) 20.09) 1.21E202)cg06685875) chr5) 4668) PCSK1) 0.06) 3.98E205) 20.08) 7.74E205)cg06691616) chr4) 2367595) SLC7A11) 0.08) 9.13E204) 20.16) 5.48E204)cg06912936) chr5) 4150) KCNIP1) 0.05) 6.35E204) 20.07) 5.60E205)cg07095945) chr15) 7697) ALPK3) 0.07) 5.05E203) 20.08) 4.86E204)cg07116919) chr7) 4358) CHRM2) 0.10) 1.41E204) 20.11) 4.78E203)cg07177437) chr1) 4583) ITLN1) 0.06) 2.05E202) 20.13) 1.18E204)cg07179075) chr1) 426808) PTBLP) 0.13) 8.51E204) 20.13) 2.96E203)cg07344172) chr1) 25666) KIF21B) 0.05) 3.11E203) 20.13) 3.34E205)cg07776698) chr7) 261468) WNT16) 0.10) 1.31E204) 20.16) 1.48E204)cg07903472) chr2) 121108) B3GALT1) 0.11) 4.24E204) 20.12) 3.75E203)cg08146841) chr6) 2483) NKAIN2) 0.08) 1.12E202) 20.14) 2.12E203)cg08250187) chr2) 2178048) 7SK) 0.06) 1.46E203) 20.12) 9.24E204)cg08278435) chr6) 1790) SLC17A2) 0.08) 1.42E204) 20.11) 4.94E204)cg08338281) chr12) 2970) WIBG) 0.08) 6.08E203) 20.11) 2.33E203)cg08468741) chr6) 3478) HLA2DMA) 0.07) 4.06E204) 20.13) 6.52E204)cg08678314) chr3) 2159494) ARHGEF26) 0.08) 1.90E202) 20.20) 2.92E205)cg08716255) chr11) 89610) FAM76B) 0.08) 3.92E204) 20.13) 9.00E205)cg08885854) chr1) 255636) 7SK) 0.08) 2.17E203) 20.16) 1.35E204)cg08991927) chr5) 21105) PPP2R2B) 0.06) 2.83E203) 20.17) 2.62E206) 247  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg09038266) chr16) 478664) MAF) 0.10) 1.19E203) 20.12) 1.82E203)cg09261794) chr6) 22227) COL11A2) 0.07) 8.70E204) 20.14) 4.23E205)cg09332996) chr1) 23363) KAZN) 0.07) 8.21E203) 20.09) 7.90E203)cg09476306) chr22) 3736) YPEL1) 0.09) 1.78E204) 20.12) 1.42E204)cg09500437) chr2) 255132) MIR4265) 0.06) 1.35E202) 20.11) 1.29E203)cg09507697) chr7) 3002) CEP41) 0.05) 7.69E203) 20.07) 1.46E202)cg09623279) chr10) 211029) TMEM72) 0.08) 1.44E203) 20.13) 2.51E205)cg09624551) chr3) 21192) AK097190) 0.04) 3.47E203) 20.07) 1.05E202)cg09648454) chr16) 22825) ABCC12) 0.04) 3.05E203) 20.06) 1.93E203)cg09664054) chr1) 2128108) HIVEP3) 0.06) 1.01E202) 20.16) 5.61E205)cg09676622) chr12) 245644) NCOR2) 0.06) 1.41E203) 20.05) 8.69E203)cg10092779) chr1) 21492) BC048114) 0.06) 6.63E204) 20.12) 1.47E205)cg10144104) chr3) 21996) CACNA2D2) 0.09) 1.46E203) 20.13) 1.26E204)cg10336671) chr7) 2341) GHRHR) 0.08) 1.54E203) 20.10) 8.84E205)cg10620457) chr1) 2422) C8B) 0.05) 1.61E202) 20.14) 9.18E207)cg10638793) chr3) 71708) MECOM) 0.07) 6.75E203) 20.11) 2.70E203)cg10791580) chr6) 2674) LOC100507203) 0.06) 1.79E204) 20.16) 7.21E205)cg11348701) chr2) 21353) GAD1) 0.08) 7.82E205) 20.14) 3.20E204)cg11377875) chr6) 2308) HMGCLL1) 0.08) 5.38E203) 20.13) 6.93E203)cg11474135) chr4) 237483) ANK2) 0.11) 3.43E203) 20.19) 4.11E204)cg12290217) chr15) 28742) AL109706) 0.04) 6.83E203) 20.06) 3.91E203)cg12387464) chr11) 21203) OR8D4) 0.06) 2.26E202) 20.20) 2.78E204)cg12493761) chr12) 2269685) PRICKLE1) 0.08) 1.01E203) 20.13) 1.36E205) 248  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg12686754) chr2) 2389) MIR216A) 0.05) 3.14E202) 20.08) 6.74E204)cg12732548) chr15) 21239) C15orf52) 0.04) 2.82E202) 20.06) 6.63E203)cg12750431) chr10) 22817) DKFZp547J0510) 0.07) 8.55E204) 20.15) 6.53E204)cg12792952) chr1) 279088) CACHD1) 0.08) 1.20E202) 20.18) 7.67E205)cg12833765) chr2) 140065) MIR3126) 0.04) 2.61E203) 20.07) 1.76E204)cg12845923) chr4) 23469) PDGFRA) 0.05) 3.59E203) 20.09) 3.26E204)cg12961926) chr17) 214400) TBCD) 0.05) 8.50E203) 20.07) 2.39E202)cg13301676) chr5) 124452) MCC) 0.06) 3.17E202) 20.17) 3.83E205)cg13345122) chr2) 187336) B3GALT1) 0.09) 1.53E202) 20.15) 6.63E204)cg13401235) chr6) 120377) LOC100130357) 0.06) 2.48E202) 20.13) 8.85E204)cg13546308) chr3) 163872) C3orf79) 0.10) 1.47E203) 20.19) 3.09E204)cg13844463) chr11) 4433) SLC1A2) 0.03) 1.29E202) 20.06) 6.34E203)cg13897222) chr1) 59852) C1orf140) 0.05) 2.14E202) 20.15) 1.22E203)cg14224444) chr6) 2235345) BRD7P3) 0.07) 5.27E204) 20.11) 2.04E203)cg14384014) chr6) 227546) U6) 0.08) 4.91E203) 20.10) 1.95E202)cg14500854) chr10) 252335) PRKG1) 0.06) 1.50E203) 20.10) 2.48E204)cg14556425) chr8) 23353) IL7) 0.08) 4.25E203) 20.16) 1.45E202)cg14578311) chr10) 214561) GTPBP4) 0.06) 3.54E203) 20.08) 1.16E202)cg14692709) chr6) 8412) COL11A2) 0.11) 2.55E203) 20.14) 1.26E203)cg14746283) chr21) 8336) pp9284) 0.09) 1.25E202) 20.13) 3.65E204)cg14851263) chr15) 21335) MYO5A) 0.08) 1.68E203) 20.12) 2.40E203)cg14938029) chr20) 967) CSTL1) 0.08) 1.70E203) 20.09) 3.42E204)cg15005526) chr2) 23223) SMYD5) 0.05) 1.40E202) 20.10) 3.49E204) 249  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg15172785) chr11) 125759) DYNC2H1) 0.09) 1.04E202) 20.14) 5.92E204)cg15410418) chr5) 2794) PDE4D) 0.06) 5.08E204) 20.12) 1.05E203)cg15492834) chr1) 2462) C1orf226) 0.05) 3.40E203) 20.10) 8.72E204)cg16056164) chr1) 217308) PCNXL2) 0.07) 7.36E203) 20.15) 1.09E205)cg16168509) chr5) 82398) PCSK1) 0.07) 2.35E203) 20.10) 6.50E204)cg16177314) chr4) 21113) GYPB) 0.08) 4.45E203) 20.16) 1.51E204)cg16243351) chr11) 211980) SLC1A2) 0.09) 1.31E203) 20.08) 7.87E203)cg16253629) chr4) 36978) ANK2) 0.04) 2.84E202) 20.09) 4.53E203)cg16326590) chr2) 1264) VWA3B) 0.06) 3.98E203) 20.04) 3.30E202)cg16423738) chr6) 2611826) IMPG1) 0.07) 1.79E204) 20.08) 3.54E203)cg16903016) chr1) 66873) 7SK) 0.05) 1.15E202) 20.14) 5.22E207)cg16926302) chr1) 77721) ST6GALNAC5) 0.07) 4.29E203) 20.15) 4.36E204)cg16964478) chr12) 21228) BC041342) 0.06) 1.99E202) 20.10) 2.55E202)cg17108958) chr1) 76803) HLX) 0.04) 4.25E202) 20.11) 1.92E204)cg17284609) chr11) 2260) SOX6) 0.08) 2.57E206) 20.11) 4.61E203)cg17295389) chr6) 14669) C6orf186) 0.05) 1.36E202) 20.10) 1.03E204)cg17357548) chr15) 218410) ATP8B4) 0.08) 4.51E204) 20.06) 3.69E202)cg17425351) chr11) 67887) 5S_rRNA) 0.07) 4.50E203) 20.16) 4.65E204)cg18018990) chr7) 1676) U6) 0.08) 8.95E203) 20.16) 1.12E204)cg18247223) chr11) 219284) TSPAN18) 0.03) 3.48E202) 20.10) 1.66E205)cg18259253) chr5) 117887) ZSWIM6) 0.10) 4.59E204) 20.12) 2.69E203)cg18285819) chr7) 2970) INMT) 0.07) 2.50E203) 20.11) 1.52E204)cg19024161) chr10) 11830) PBLD) 0.05) 1.22E202) 20.12) 1.28E204) 250  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg19118951) chr21) 22093) LINC00310) 0.05) 8.86E203) 20.06) 9.27E203)cg19143211) chr6) 78972) AL832096) 0.11) 1.32E204) 20.14) 2.66E203)cg19183252) chr4) 59729) BC038750) 0.06) 4.56E203) 20.12) 1.14E204)cg19192065) chr10) 36126) LOC100188947) 0.04) 4.40E202) 20.12) 2.52E205)cg19578175) chr11) 459914) SOX6) 0.05) 1.39E203) 20.07) 6.80E204)cg19760250) chr17) 15) C17orf98) 0.15) 1.76E203) 20.07) 2.83E203)cg19786983) chr15) 68936) AK094929) 0.05) 1.51E202) 20.06) 1.30E202)cg19883388) chr8) 233533) AZIN1) 0.07) 1.02E202) 20.08) 3.03E202)cg19937056) chr3) 2919) KCNMB2) 0.07) 1.63E202) 20.21) 6.72E204)cg20054786) chr4) 1046) TLR1) 0.06) 3.57E203) 20.14) 2.42E206)cg20090782) chr18) 62854) SETBP1) 0.05) 3.75E203) 20.11) 3.55E205)cg20092728) chr11) 507) SLC5A12) 0.06) 9.17E203) 20.12) 1.76E202)cg20389635) chr12) 2840) PTHLH) 0.07) 2.52E202) 20.15) 1.28E205)cg20497304) chr2) 248827) ATP5G3) 0.05) 9.67E203) 20.13) 1.05E205)cg20570458) chr3) 233738) KALRN) 0.06) 1.35E203) 20.06) 5.01E204)cg20782816) chr3) 3608) CAMK1) 0.05) 9.70E204) 20.09) 2.56E204)cg20822693) chr2) 2684) SLC16A14) 0.04) 7.02E203) 20.10) 2.15E204)cg20991819) chr1) 97472) U3) 0.06) 1.78E204) 20.10) 8.97E205)cg21182526) chr2) 13978) MIR548N) 0.08) 6.11E203) 20.18) 3.02E204)cg21283167) chr1) 84557) FOXD2) 0.06) 8.87E204) 20.06) 2.92E202)cg21760128) chr3) 22471) LARS2) 0.03) 2.68E202) 20.07) 5.35E204)cg21775668) chr6) 21328) LOC100507203) 0.07) 5.33E203) 20.18) 9.64E205)cg21837826) chr1) 4434) FAM163A) 0.05) 3.38E203) 20.11) 1.77E204) 251  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg21977522) chr3) 20133) U6atac) 0.10) 1.23E203) 20.17) 1.41E203)cg22061561) chr11) 3638) TMEM86A) 0.06) 4.44E202) 20.14) 1.12E205)cg22245446) chr2) 149101) DIRC3) 0.06) 8.61E204) 20.11) 8.09E206)cg22303227) chr17) 2169) AOC4) 0.03) 3.79E202) 20.07) 1.92E203)cg22497133) chr7) 2668995) DQ599872) 0.09) 2.88E204) 20.09) 1.21E202)cg22520372) chr4) 67) NPY5R) 0.07) 5.65E204) 20.11) 2.34E204)cg22671342) chr1) 6552) IL24) 0.04) 1.72E203) 20.07) 1.55E204)cg22799757) chr7) 22666) SCRN1) 0.07) 1.71E204) 20.09) 1.48E203)cg22828602) chr7) 629) RNF133) 0.07) 6.14E203) 20.19) 1.78E205)cg23101885) chr2) 3681) RAD51AP2) 0.08) 2.22E203) 20.08) 1.23E202)cg23104436) chr11) 2335308) SnoMBII_202) 0.08) 2.01E205) 20.11) 1.37E205)cg23128495) chr12) 279328) NUAK1) 0.08) 6.33E204) 20.14) 3.49E205)cg23255934) chr2) 2194088) TRNA_Pseudo) 0.10) 8.53E205) 20.11) 8.66E203)cg23309843) chr5) 2881) CTXN3) 0.08) 1.49E203) 20.16) 6.49E206)cg23330212) chr11) 8499) BDNF) 0.05) 1.62E202) 20.08) 3.28E202)cg23400222) chr5) 2588206) ZNF608) 0.09) 8.52E204) 20.12) 3.38E203)cg23516370) chr7) 2727456) DQ599872) 0.12) 1.28E204) 20.17) 7.13E204)cg23697796) chr10) 36243) LOC100188947) 0.05) 1.21E202) 20.17) 2.72E206)cg23831864) chr4) 2282236) U6) 0.07) 4.51E204) 20.14) 9.55E204)cg23931150) chr10) 40184) DKK1) 0.10) 3.01E205) 20.14) 1.56E203)cg24233594) chr10) 243064) AX747408) 0.04) 2.75E202) 20.05) 1.67E202)cg24340655) chr1) 282770) RGS4) 0.05) 9.52E204) 20.15) 2.02E205)cg24700316) chr6) 63065) BACH2) 0.08) 1.37E205) 20.12) 1.97E204) 252  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg24882192) chr2) 211232) MYO7B) 0.08) 9.58E203) 20.10) 1.55E204)cg24995347) chr2) 803) SP110) 0.06) 5.52E203) 20.07) 3.75E202)cg25099516) chr1) 173749) LOC388630) 0.05) 2.07E202) 20.14) 2.47E205)cg25103160) chr12) 2123739) SNORA70) 0.04) 5.49E204) 20.08) 1.08E203)cg25212701) chr1) 145) C1orf226) 0.06) 2.49E202) 20.10) 1.20E203)cg25390063) chr12) 233405) AK128707) 0.05) 1.05E202) 20.13) 2.76E205)cg25429446) chr4) 99807) ANK2) 0.08) 2.26E203) 20.17) 1.32E205)cg25469314) chr1) 2162) SLC5A9) 0.06) 4.51E204) 20.10) 9.48E204)cg25783987) chr7) 11407) SDK1) 0.07) 8.39E203) 20.08) 3.80E202)cg25989745) chr9) 259) TYRP1) 0.07) 1.82E203) 20.13) 2.31E204)cg26020161) chr1) 2763) GSTM3) 0.06) 6.11E205) 20.10) 2.07E203)cg26276667) chr10) 23452) NUDT9P1) 0.06) 1.02E203) 20.13) 4.04E206)cg26331247) chr20) 612) C20orf203) 0.05) 2.53E202) 20.10) 9.92E204)cg26365254) chr1) 210944) KIRREL) 0.06) 1.58E202) 20.11) 6.23E204)cg26366616) chr8) 1845) PDLIM2) 0.03) 4.85E202) 20.06) 1.42E202)cg26378984) chr11) 224329) Mir_652) 0.05) 4.59E203) 20.08) 2.14E203)cg26548883) chr17) 2815) CCL142CCL15) 0.07) 2.15E203) 20.14) 4.16E204)cg26851803) chr12) 267346) 5S_rRNA) 0.07) 6.01E204) 20.10) 2.76E204)cg26931063) chr6) 166) OPN5) 0.07) 2.45E203) 20.10) 3.06E202)cg27223882) chr2) 89976) RPRM) 0.06) 3.69E203) 20.15) 2.04E203)cg27321175) chr10) 72761) ITGA8) 0.04) 3.87E202) 20.09) 1.21E202)cg27423027) chr6) 2744) LOC100507203) 0.06) 7.46E204) 20.16) 2.75E205)cg27442676) chr4) 17665) SLC4A4) 0.12) 7.35E206) 20.17) 3.67E204) 253  Probe& Chromosome&Closest&TSS&Gene&Name& Distance_closest_TSS&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Discovery&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&Δβ&Validation&Cohort&EOPE&subcluster&1&vs&EOPE&subcluster&2&p?value&Validation&Cohort&cg27492725) chr2) 221964) PDE1A) 0.07) 7.93E205) 20.12) 2.83E204)cg27502580) chr3) 132454) DD413694) 0.06) 1.95E202) 20.17) 5.13E204)cg27589988) chr1) 230993) KCNK2) 0.07) 4.85E203) 20.18) 5.69E205) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0361940/manifest

Comment

Related Items