UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Application of SCODA to forensic exhibits Mai, Laura 2011

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

Item Metadata

Download

Media
24-ubc_2011_fall_mai_laura.pdf [ 9.3MB ]
Metadata
JSON: 24-1.0105145.json
JSON-LD: 24-1.0105145-ld.json
RDF/XML (Pretty): 24-1.0105145-rdf.xml
RDF/JSON: 24-1.0105145-rdf.json
Turtle: 24-1.0105145-turtle.txt
N-Triples: 24-1.0105145-rdf-ntriples.txt
Original Record: 24-1.0105145-source.json
Full Text
24-1.0105145-fulltext.txt
Citation
24-1.0105145.ris

Full Text

APPLICATION OF SCODA TO FORENSIC EXHIBITS by Laura Mai B.Sc., The University of British Columbia, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Genome Science and Technology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August 2011  © Laura Mai, 2011  Abstract Since law enforcement first used DNA evidence to solve crimes in 1985, DNA typing and the comparison of STR (Short Tandem Repeat) profiles have become the accepted gold standard in forensic science for identification. Although common DNA extraction methods such as phenol-chloroform extraction and silica binding matrices are often effective, a small percentage of samples fail to yield a profile. Some types of samples, for example bones or bloodstains on concrete, are particularly challenging due to environmental degradation of the DNA and high concentrations of PCR inhibitors relative to the amount of available DNA. This ultimately makes obtaining a DNA profile very difficult and sometimes impossible for this class of samples. This thesis introduces SCODA (Synchronous Coefficient of Drag Alteration) as a novel electrophoretic nucleic acid clean-up method for purifying trace amounts of DNA from PCR inhibited forensic samples. Unlike other extraction techniques that rely on the chemical properties of DNA, SCODA takes advantage of the physical properties unique to long and charged molecules such as nucleic acids to selectively concentrate them in an agarose gel matrix. SCODA was compared to other extraction and purification methods to determine the best method for obtaining DNA profiles from difficult forensic samples. Samples that were examined included bones, bloodstains on concrete and two actual forensic exhibits. DNA yield and PCR inhibition were assessed by quantitative PCR and STR analysis was conducted to ensure that profiles could be obtained. In summary, it has been found that SCODA is optimal for highly inhibited samples that mask the presence of DNA, as in some cases of bone, and that specific protocols can be designed to further improve SCODA to outperform other extraction methods, which are more difficult to customize.  
 ii
  Preface The work presented in Chapter 3 and 4, where human bone and blood were used, was approved by the UBC Clinical Research Ethics Board; the certificate number is H0902691. A second certificate was granted by the Canadian Blood Services Research Ethics Board to obtain human whole blood, which was used in Chapter 4; the reference number of this certificate is 2010.005. Chapter 5 was a concerted effort by myself, Dr. David Sweet O.C., and Diane Fairley from the Bureau of Legal Dentistry (BOLD) Lab. It was Dr. Sweet’s idea to submerse the object from section 5.1 in lysis buffer, Diane carried out the extraction with phenol-chloroform, the DNA quantification, and the STR profiling. I conducted SCODA purification and applied the speed-vac. In both cases reported in Chapter 5, Diane and I both performed the STR analysis. Guidance for statistical calculations were provided by Matt Wiggin from Boreal Genomics Inc.  
 iii
  Table of Contents Abstract ..............................................................................................................................ii Preface .............................................................................................................................. iii Table of Contents..............................................................................................................iv List of Tables.....................................................................................................................vi List of Figures ..................................................................................................................vii List of Abbreviations..................................................................................................... viii Glossary .............................................................................................................................ix Acknowledgements ...........................................................................................................xi Chapter 1 – Introduction ..................................................................................................1 1.1 Forensic Science........................................................................................................1 1.2 Genetic Diversity.......................................................................................................2 1.3 DNA as a Tool in Forensic Biology ..........................................................................3 1.3.1 History ................................................................................................................3 1.3.2 Short Tandem Repeats........................................................................................6 1.3.3 DNA Profiling: Modus Operandi.......................................................................9 1.4 DNA Extraction.......................................................................................................12 1.4.1 Phenol-Chloroform Extraction .........................................................................12 1.4.2 Silica-Based Extraction ....................................................................................13 1.4.3 SCODA Technology ........................................................................................14 1.5 PCR in Forensic DNA Typing ................................................................................15 1.5.1 Quantification ..................................................................................................15 1.5.2 STR Amplification ...........................................................................................16 1.6 STR Analysis...........................................................................................................17 1.6.1 Capillary Electrophoresis .................................................................................17 1.6.2 Genotyping .......................................................................................................18 1.6.3 Interpretation ....................................................................................................18 1.7 Thesis Objectives.....................................................................................................19 Chapter 2 – SCODA........................................................................................................21 2.1 SCODA Principle ....................................................................................................21 2.2 Process.....................................................................................................................23 2.2.1 Qt Injection.......................................................................................................23 2.2.2 Washing and SCODA Focusing.......................................................................24 2.3 Instruments ..............................................................................................................25  
 iv
  Chapter 3 – Bone Samples ..............................................................................................27 3.1 Introduction .............................................................................................................27 3.2 Methods and Materials ............................................................................................29 3.3 Results .....................................................................................................................31 3.4 Discussion................................................................................................................39 Chapter 4 – Bloodstains on Concrete ............................................................................44 4.1 Introduction .............................................................................................................44 4.2 Methods and Materials ............................................................................................45 4.3 Results .....................................................................................................................48 4.4 Discussion................................................................................................................56 Chapter 5 – Casework.....................................................................................................59 5.1 A Murder Weapon...................................................................................................59 5.1.1 Introduction ......................................................................................................59 5.1.2 Methods and Materials .....................................................................................60 5.1.3 Results ..............................................................................................................61 5.1.4 Discussion.........................................................................................................64 5.2 Human Remains ......................................................................................................65 5.2.1 Introduction ......................................................................................................65 5.2.2 Materials and Methods .....................................................................................65 5.2.3 Results ..............................................................................................................66 5.2.4 Discussion.........................................................................................................69 Chapter 6 – Conclusions .................................................................................................71 References ........................................................................................................................73 Appendix A – Methods and Materials...........................................................................77  
 v
  List of Tables Table 3.1 – Summary of processed bones from the BCCS. ..............................................33 Table A.1 – SCODA Run Conditions. ..............................................................................79  
 vi
  List of Figures Figure 1.1 – Two forms of DNA variation..........................................................................3 Figure 1.2 – Illustration of RFLP analysis. .........................................................................5 Figure 1.3 – VNTR and STR markers.................................................................................7 Figure 1.4 – Inside view of the ABI Prism 310 Genetic Analyzer....................................11 Figure 1.5 – Sample DNA profile. ....................................................................................12 Figure 1.6 – Demonstration of SCODA concentration. ....................................................15 Figure 2.1 – Top view of a disposable cartridge. ..............................................................22 Figure 2.2 – SCODA injection and concentration.............................................................23 Figure 2.3 – Electrophoretic washing of inhibitors. ..........................................................24 Figure 2.4 – Prototype SCODA instruments.....................................................................26 Figure 3.1 – Cut Sagittal-section of a femur: sample BB7................................................29 Figure 3.2 – Quantifiler™ Human DNA Quantification Results for BB7. ........................34 Figure 3.3 – STR profile of BB7 as purified by SCODA. ................................................35 Figure 3.4 – STR profile of BB7 purified by SCODA with increased injection time. .....36 Figure 3.5 – Quantifiler™ Human DNA Quantification Results for BB9. ........................38 Figure 3.6 – Quantifiler™ results for BB9 extraction controls.........................................39 Figure 3.7 – Slab gel electrophoresis of DNA extracts from BB7....................................40 Figure 4.1 – Collection of bloodstains on concrete...........................................................46 Figure 4.2 – Comparison of standard and EDTA lysis with PC or SCODA extraction....49 Figure 4.3 – DNA extraction from 10µl bloodstains on concrete. ....................................51 Figure 4.4 – Summary of swabbed and chipped 10µl BSC...............................................52 Figure 4.5 – Summary of swabbed and chipped 1µl BSC.................................................53 Figure 4.6 – STR Analysis of Chipped 1µl BSC...............................................................55 Figure 5.1 – Illustration of the object processed twice with two different volumes. ........60 Figure 5.2 – Quantifiler™ Human DNA Quantification Results. ......................................62 Figure 5.3 – STR Profile Obtained from SCODA-spVac .................................................63 Figure 5.4 – Quantifiler™ Human DNA Quantification Results. ......................................67 Figure 5.5 – PC extracted STR Profile of Q......................................................................68 Figure 5.6 – SCODA extracted STR Profile of Q. ............................................................69  
 vii
  List of Abbreviations BCCS – British Columbia Coroners Service CE – Capillary Electrophoresis CODIS – Combined DNA Indexing System CSI – Crime Scene Investigation EDTA – Ethylenediaminetetraacetic acid FBI – Federal Bureau of Investigation FAD H2O – Filtered Autoclaved Distilled water HsBB – High-speed Bread Board SCODA instrument prototype IPC – Internal Positive Control PC – Phenol:chloroform:isoamyl alcohol DNA extraction followed by ultrafiltration PCR – Polymerase Chain Reaction RCMP – Royal Canadian Mounted Police RFLP – Restriction Fragment Length Polymorphism qPCR – Quantitative Polymerase Chain Reaction SCODA – Synchronous Coefficient of Drag Alteration SDS – Sodium dodecyl sulphate STR – Short Tandem Repeat TBE – Tris-Borate-Ethylenediaminetetraacetic acid VNTR – Variable Number Tandem Repeat  
viii
  Glossary Allelic dropout – An event that occurs during PCR amplification where a heterozygous genotype appears to be homozygous due to preferential amplification of one allele over the other, resulting in a bias representation. This is typically seen in heavily degraded samples or in reactions containing low template DNA. CΤ – the PCR cycle number at which the fluorescence signal is greater than the threshold arbitrarily selected to be above baseline during the exponential phase of qPCR. k – coefficient of nonlinearity applied to long, charged polymers to describe their motion in a gel under the influence of an electric field. Discriminatory power – a statistical calculation using population allele frequencies to measure the potential power of a system of selected genetic markers, to differentiate between any two people chosen at random. Locus – the specific chromosomal location of a gene or DNA sequence. Processivity – the average number of nucleotides that can be added onto a growing nucleic acid strand by a polymerase enzyme before it dissociates. Standard 9947A – DNA standard of a known profile that is included in all Applied Biosystems and Promega STR amplification kits as a positive control. Small volume samples – referring to casework samples that require less than 1ml lysis volumes for extraction including swabs. Taq polymerase – a thermostable polymerase isolated from the bacterial species Thermus aquaticus that inhabits deep thermal vents.  
 ix
  Touch-DNA – DNA that has been transferred onto surfaces having been touched by an individual(s).  Qt – Charge threshold is a time integral of the current and estimate of the amount of molecular charge that can be accepted by a gel during SCODA injection; it allows for injection to be reliable and efficient.  
 x
  Acknowledgements 
  Exceptionally blessed, these are the words that come to mind when I reflect on my academic journey… First, I would like to thank my co-supervisor, Dr. Andre Marziali, whose ingenuity and passion for science is incredibly admirable. I thank him for giving me my first job as a co-op student, inspiring me to do my masters, and the opportunity to fulfill my career goal in conducting research in forensic science. I am extremely grateful for the support he has given me. I would also like to thank Dr. David Sweet O.C., also my co-supervisor, who I have the utmost respect for and am extremely honoured to have worked with. His vast knowledge, and excellence in forensic science was crucial in making this research significant. I have learnt so much from him and cannot thank him enough for accepting me as his student. Next, I would like to thank Dr. Joel Pel, my “unofficial supervisor”, for his exceptional insights on SCODA and countless hours of guidance on experimental design, which have helped shaped this work. Similarly, thanks to Dr. Jason Maydan, Hau-Ling Poon, Sweta Rajan, Tim Smith, Jaryn Perkins, Leyla Tabanfar, the rest of the Boreal Genomics crew and Diane Fairley from the BOLD Lab, for their thoughts, support and making my 2-yr experience truly delightful. Thanks also to Steven Fonseca and Bill Inkster from the BC Coroners Service for supplying the bones for my research, and Hiron Poon from the RCMP Forensic Lab for his enthusiasm and expertise. Finally, I owe a big, big thanks to my family and friends for their unconditional love and support for me through the years. I could not have done this without you!
  
 xi
  Chapter 1 – Introduction 1.1 Forensic Science It is difficult to know the exact sequence of events at a crime scene. Perpetrators are rarely forthcoming about their crimes, victims may not have a chance to tell their stories, and neither of them will necessarily present events correctly or objectively. Forensic science is the application of science to a court of law; it allows police officers to piece together solid evidence or clues that lead to the truth 1. Like other scientific disciplines, forensic science involves employing reliable, dependable, and predictable methodologies along with impeccable attention to detail in order not to leave any possibilities untested or any gaps in credibility. Tools applied to analyze evidence at crime scenes must give correct and repeatable results to ensure that the evidence-based story presented in court is an accurate representation of what really happened: justice and the liberty, or perhaps life, of the accused are at stake. If the evidence is not reliable and changes over time it cannot be concluded without a doubt that the suspect was responsible for a crime. DNA evidence is considered one of the finest pieces of evidence that could be left at a crime scene. This is because it is unique to each individual (with the exception of identical twins), does not change over time, and is prohibitively difficult to manipulate. DNA is left at crime scenes through transference explained by the Locard Exchange Principle, which states that when two objects come into contact with each other, each object will leave trace amounts of material onto the other 2. Therefore, finding an individual’s DNA at a crime scene is compelling evidence that he or she was present at the scene.  
 1
  1.2 Genetic diversity 
  Approximately 99.5% of the human genome is identical in all humans 1. The  remaining 0.5% gives each human his or her identity in a forensic context 3. Much of this tiny fraction that is unique in every individual, including the regions of greatest interest to forensic scientists, is found within the non-coding regions of DNA. Possible variants of a gene are called alleles 4. Normally, an individual will have only two alleles for each gene, one from each of his or her parents. During the formation of a zygote, DNA from each parent undergoes random recombination. At each gene and locus, an offspring will acquire either allele from their mother and either allele from their father from the four possible alleles between the two parents. As a result, two siblings unless they are identical twins will have distinct genomes as this process of acquiring alleles is completely random. Although each individual carries two alleles for any given locus that is not to say that there are only two possible alleles in the entire population. The existence of multiple alleles for a locus is called polymorphism. There are two primary types of polymorphisms found in DNA: sequence polymorphisms and length polymorphisms 1. Sequence polymorphisms are like spelling the same word two different ways and length polymorphisms are like having that word repeated a variable number of times (Figure 1.1).  
 2
  Figure 1.1 – Two forms of DNA variation. Image reprinted with permission from Butler 2005.  1.3 DNA as a Tool in Forensic Biology 1.3.1 History The first criminal case ever to be solved with DNA evidence was in 1984 by Sir Alec Jeffreys, a professor at the University of Leicester 1. The case involved the rape and murder of two young victims whose bodies were found in close proximity to each other. Vaginal swabs from the victims yielded DNA from semen, which Jeffreys was able to use to conclude that the victims were raped by the same perpetrator. Having canvassed the surrounding areas to look for potential leads and request DNA samples, the police narrowed in on one suspect who had suspiciously asked a colleague to submit DNA in place of his own. Jeffreys compared the DNA from this suspect to the DNA found on the victims and established a positive identification. Jeffreys had discovered that there were loci in the DNA that contained islands of tandem repeat sequences, called satellite regions 1. He noted that the number of repeating 
 3
  units was highly variable from one individual to another, as were the sequences from one locus to the next; today they are referred to as VNTRs, or variable number of tandem repeats 1. To visualize these VNTRs, restriction enzymes were used to cut the DNA at known locations into lengths that differed according to the lengths of the VNTRs. These fragments were separated by size via gel electrophoresis, transferred to a nylon membrane and then hybridized with radiolabeled multilocus minisatellite probes to detect many loci at a time. This pattern, which Jeffreys and colleagues coined a DNA profile or DNA fingerprint, resembled a barcode and was hypothesized to be distinct in each individual. Today, this type of DNA analysis is referred to as RFLP analysis. Since multilocus probes detect many different loci on different chromosomes from a single human DNA source, samples containing DNA from two different individuals proved difficult to analyze; the barcode pattern was no longer a result from one individual 1. For this reason, single locus probes were developed to target specific minisatellites in North America. In this case, the pattern from each targeted locus either yielded one or two bands, representing the number of sequence repeats and if that individual was homozygous or heterozygous for that particular locus (Figure 1.2). Because the use of a single locus probe produces only one or two bands, the likelihood that two individuals would have the same pattern from a single locus is relatively high, so forensic analysts often used many single locus probes collectively 1,3. As a result, the probability that two people have the same pattern at every analyzed locus was highly unlikely. This technique was also advantageous in paternity tests, where DNA of the offspring is compared to the father since one could tell which band(s) were inherited from the father. RFLP analysis using single locus probes was the first type of DNA analysis officially accepted for forensic use by United States 5. In Canada, the RCMP first used DNA analysis in a sexual assault case in 19896.  
 4
  Figure 1.2 – Illustration of RFLP analysis. Site-specific restriction enzymes are used to cut DNA, resulting in products of varying lengths between different individuals (A,B,C). Digests are separated by slab gel electrophoresis, producing a banding pattern unique between individuals. Although the advent of RFLP analysis using multilocus and single locus probes revolutionized forensic analysis, the techniques were labourious and required an abundant source of high quality DNA 3. Even though the transition to single locus probes reduced the amount of required DNA from 100ng to 10ng, the DNA needed to be of high quality. This is frequently challenging especially in forensic cases where DNA is often subjected to harsh environments resulting in degradation or low DNA yields. Fortunately, these limitations were overcome with the introduction of PCR, which consequently lead to STR analysis.  
 5
  1.3.2 Short Tandem Repeats Like the VNTRs used for RFLP analysis, STRs are used for STR analysis 1,3. The major difference between VNTRs and STRs is that the former consists of a core repeat unit of approximately 10-100 bases in length, whereas the latter has a repeat unit consisting of only 2-6bp (Figure 1.3). Including the highly variable number of tandem repeats, a VNTR allele can range from ~ 400-1 000bp where an STR allele will range from ~100-400bp 1. The small size of STRs appeal to forensic applications since degraded DNA is more commonly encountered than non-degraded. In this aspect, the probability of DNA being sheared in an STR marker, which would render it unusable for generating a profile, is much lower than in a VNTR marker 7. STRs have also become the preferred DNA repeat markers since they can easily be amplified by PCR without running into differential amplification issues 1,3,8. For example, in a heterozygous individual who may have two alleles of extremely different lengths at one marker, the shorter allele will amplify more favourably since amplification efficiency decreases as amplicon length increases giving a biased representation of the DNA 5. Because STRs all have similar sized units, preferrential amplification is not as problematic as it is for RFLP analysis. The number of alleles or tandem repeats present in the population can range from 5-20 depending on the population (Figure 1.3b). Therefore, the fact that the number of repeating units in STR markers is highly variable among individuals makes them just as effective if not more than RFLPs in discriminating between individuals.  
 6
  Figure 1.3 – VNTR and STR markers. (a) Schematic comparison of VNTRs and STRs. (b) Two possible alleles for the D5S818 locus. Image reprinted with permission from Butler, 2005.  It has been found that some locations in human genomic DNA are more likely to have polymorphisms between individuals than others 5. Population variation studies on a number of STR candidates have prompted a consensus set of variable loci to be established as markers by the forensic community for human identification 5. After seeing STR analysis successfully deployed in the U.K., the FBI launched a project to establish a set of core STR loci for its forensic cases.  
 7
  Candidate STR markers are evaluated based on several characteristics. Tetranucleotide repeats as opposed to dinucleotide and trinucleotide repeats are preferred since they have a lower stutter percentage 1,3,9. Stutter occurs due to Taq polymerase slippage during PCR and results in artifacts that are one or two repeat units less than the true product. A larger stutter percentage makes it difficult to analyze mixed samples. Penta- and hexanucleotide repeats are preferred as well but are very rare. Shorter fragments are desired since they have a high rate of amplification over longer fragments in heavily degraded samples. Low mutation rates are required for continuity purposes. Location on separate chromosomal locations is necessary to ensure the STRs are not linked. Selecting STRs on different chromosomes allows forensic analysts to take advantage of the product rule, which states that the probability of having mutually exclusive events occur simultaneously is the product of the probabilities of each event, intensifying the technology’s discriminatory power. Today, 13 markers form the core STR loci used in forensic DNA analysis by the FBI and RCMP; combined, they are referred to as CODIS. The names of the loci are CSF1PO, FGA, TH01, TPOX, vWA, D3S1358, D7S818, D7S820, D8S1179, D13S317, D16S539, D18S51, and D21S11 3,5. The 13 CODIS markers are standardized and used to construct Canada’s National DNA Data Bank, the U.S. National DNA Database, as well as other criminal justice databases around the world, which permits sharing DNA profiles across many jurisdictions. Together, the 13 STR markers provide a discriminatory power exceeding a minimum of 1 in a billion 3. This value is how one can be certain in identifying individuals.  
 8
  1.3.3 DNA Profiling: Modus Operandi Obtaining a DNA profile from a crime scene is not as easy as it looks on CSI. A good source of nuclear DNA from the crime scene must first be identified and collected. Common sources of DNA include biological fluids like blood, semen, and saliva, or shed tissue like hair roots, bones, and teeth. Although the surface of the skin contains few or no nuclei, DNA can be deposited onto the surface of the skin through perspiration (sweat and oil secretion), so shed skin cells can be a potential source of DNA too 1. When a crime scene has been cleaned up or is in the natural environment, it becomes a challenge in itself to sift through and find something that could yield DNA. For this reason presumptive tests, which can help indicate if a source is a biological fluid, have been developed to help with the searching process. Once a source of DNA has been found and collected the next step to obtaining a DNA profile is to extract the molecule from the cells that encase it. There are many methods used to extract DNA. Common extraction methods include a lysis step, often involving a detergent or a strong base that will lyse and break open the cells to release DNA into solution and Proteinase K, an enzyme that aggressively cleaves proteins associated with DNA or that degrade DNA, like nucleases. The second step involves purifying DNA from cellular structures and proteins; this can be done by adding an organic solution that separates the solution into two phases to pull only DNA into the non-organic phase, or by binding DNA to a substrate, a membrane or beads, which can then be washed to remove other cellular components. The third and final step of extraction is to concentrate the DNA by suspending it in a small volume. Following extraction, DNA is quantified to assess the amount of DNA recovered from the sample. Although the usability of a sample for forensic casework is determined by its ability to generate an STR profile and not necessarily the absolute amount of DNA recovered, DNA is first quantified to estimate the appropriate amount of DNA extract to be added to the amplification process to avoid signal saturation. Today, a PCR-based method called quantitative PCR (qPCR), also known as real-time PCR, is used to 
 9
  accurately quantify DNA by incorporating fluorescent dye(s) into the system. The dyes are excited by a light source at the end of each cycle and emit a fluorescent signal detected by a CCD camera that is relayed to a computer that measures the intensity. At the exponential phase of PCR, the intensity of fluorescent signal is directly proportional the amount of DNA in the sample. The next step is to amplify the DNA using forward and reverse primer pairs designed to target the selected STR markers. Today, commercially available multiplex STR kits, designed to target multiple loci at a time, are widely used in forensic labs since they are reliable and easy to use. Fluorescent dyes are incorporated into the amplicons during the amplification process. The PCR product from each STR locus is uniquely identified by a combination of the size of the amplicon and which fluorophore is incorporated during amplification. For an optimal STR profile to be developed, many of these commercial STR kits specify that DNA be diluted to a range of 1-1.25ng. Adding too much template DNA could result in a saturated profile with nonspecific peaks while adding too little can result in allelic dropout and fluorescence signals below threshold. Finally, because the end PCR products contain various lengths of amplified alleles and template, a separation and visualization step is needed to resolve the alleles that make up a DNA profile. One of the most widely used systems in STR analysis is the genetic analyzer ABI Prism 310. It is a fully automated CE system that consists of a single capillary, laser excitation source, fluorescence detectors and a computer that collects and processes data (Figure 1.4). Each commercialized STR kit comes with an allelic ladder that contains all known alleles at each locus; this is always included in every CE setup since sample alleles need to be compared to the ladder for proper sizing and naming. The end product of this system is a DNA profile that is displayed on an electropherogram where alleles appear as coloured peaks, which correspond to the fluorescent dye they were labeled with (Figure 1.5). Loci are displayed according to allelic size.  
 10
  Figure 1.4 – Inside view of the ABI Prism 310 Genetic Analyzer. The 310 Genetic Analyzer system provides an all-in-one solution for CE, fluorescence detection and data collection and analysis.  
 11
  Figure 1.5 – Sample DNA profile. X-axis is DNA fragment length in bp; Y-axis is signal intensity in relative fluorescent units (RFUs); alleles are peaks within the corresponding loci. Red markers show the range of alleles within the locus. Grey vertical bars show where alleles are expected to appear.  1.4 DNA Extraction  1.4.1 Phenol-Chloroform Extraction Obtaining a DNA profile from a sample depends on efficiently extracting the DNA and excluding inhibitors that can interfere with the sensitive PCR procedures.  
 12
  Phenol-chloroform organic extraction is most commonly used especially on difficult forensic samples since it reliably produces high purity with good yield 10. Organic extraction works on the principle of differential solubility 11. When a typical mixture of phenol:chloroform:isoamyl alcohol at a ratio of 25:24:1 is added to a sample that has been lysed with detergent and Proteinase K, DNA will solubilize into the aqueous phase. Chloroform, although it has similar solvent properties to phenol, is added to help stabilize the boundary between the two phases. Isoamyl alcohol prevents frothing between the layers as they are mixed and contributes to boundary stabilization 11. Commercial phenol mixtures are buffer-saturated to neutralize the naturally acidic chloroform; this neutralization is necessary to prevent DNA, which is also slightly acidic, from migrating to the organic layer 11. Following centrifugation to separate the layers, lipids, polysaccharides and other cellular material remain in the organic phase. Proteins, denatured by phenol, precipitate and form a layer at the interface. Organic extraction is often repeated a few times in order to remove as much cellular debris as possible. With small volume samples, a phase-lock gel—a microfuge tube that contains a wax pellet— is often used to facilitate the collection of the aqueous phase and minimize contamination of phenol and interface molecules. To prevent excess phenol from carrying through downstream processes, a final extraction is performed using a solvent like N-butanol or chloroform alone.  1.4.2 Silica-Based Extraction DNA extractions employing silica beads or membranes rely on the ability of silica to adsorb nucleic acids to its surface in the presence of chaotropic salts, which remove water from hydrated molecules in solution or under low pH conditions 11. Molecules with dissimilar chemical affinities to DNA do not bind to the matrix and pass through. After a few wash steps to remove any remaining particulates, DNA is eluted using a neutral pH buffer or water.  
 13
  Manufacturers such as Promega produce paramagnetic beads coated with silica resins that can be pulled down by applying an external magnet to separate the beads from solution. Bead methods do not require centrifugation between wash steps and are easily automated for high-throughput applications 12. Others have developed silica membrane spin columns that function in a similar manner to bind, wash and elute DNA. Some disadvantages to using silica-based methods are that they can carry through PCR inhibitors that have similar binding properties to DNA and in small amounts, may fail to elute DNA from the silica 13. Also, given the large presence of inhibitors and cellular debris in some samples including bone, silica membranes may physically clog and prevent the sample from flowing through. Silica-based extraction methods are nonetheless the method of choice for small volume casework samples in many forensic laboratories due to their relative ease of use.  1.4.3 SCODA Technology SCODA is a novel nucleic acid extraction and purification technique that operates on the principles of electrophoresis, a technique for molecular separation. Rather than rely on chemical affinities to substrates or solubility in organic solutions to achieve extraction, SCODA depends on the physical aspects of DNA. Nucleic acids are known to exhibit complex electrophoretic behaviour when they migrate through a separation medium such as an agarose gel due in part to their long contour lengths and high linear charge density. In particular, the drift velocity of DNA migrating through a gel under an electric field responds nonlinearly to the strength of that field while small molecules obey a linear velocity-field relation 14. In 2005, Marziali et al. exploited this nonlinearity to show that two synchronously rotating electric fields, each of which apply a timeaveraged zero net field, could induce net drift of DNA molecules. When the rotating fields are positioned around an agarose gel, as illustrated in Figure 1.6, nucleic acids will migrate to and focus at the centre of the electrodes and gel. Other biomolecules that do not exhibit this nonlinearity do not focus and can even be expunged from the gel by applying a small bias field. SCODA can be used to separate and purify DNA from  
 14
  inhibited samples with minimal sample handling, potentially simplifying labourious workflows.  Figure 1.6 – Demonstration of SCODA concentration. Image A, shows the placement of electrodes (black bars) around a concentration gel (centre). Images B-H is a time‐lapse sequence showing concentration of SYBR® Green I stained pUC19 DNA (2.7kb) from a homogeneous distribution of 0.2ng/µl of DNA in 1% low melting point agarose and 0.25xTBE. Images are taken at 10min intervals, for a total run time of 60min. The concentration of DNA in the focused spot is estimated to be 100200ng/µl. Image reprinted from with permission from J. Pel et al. 2009.  1.5 PCR in Forensic DNA Typing 
 1.5.1 Quantification Applied Biosystems’ Quantifiler™ Human Quantification Kit is the standard qPCR assay used by the forensic community for quantifying total human DNA. It is a  
 15
  TaqMan® assay that specifically targets a 62bp region of the human telomerase reverse transcriptase locus found on chromosome five 15,16. TaqMan® PCR systems use hydrolysis probes designed to be complementary to an internal region within the desired template to quantify amplified DNA. This method of labeling enhances signal specificity, limiting detection to only the desired template as opposed to other methods that detect all double-stranded DNA. Absolute quantification is achieved by comparing samples to a DNA standard, included in the kit, which is serially diluted to known concentrations. The Quantifiler™ kit co-amplifies a synthetic DNA internal positive control (IPC) present in the reaction mix to assess for inhibition 15-17. Similar amplification times are expected for each reaction since the same amount of IPC is deposited into every reaction well. IPC amplification is delayed if DNA or reaction components are sequestered in cases where inhibitors are present. Therefore, IPC results are assessed to ensure that a negative result is not merely caused by inhibition and can also be used to indicate whether reagents and instruments are performing properly 16.  1.5.2 STR Amplification There are several commercially available multiplex PCR kits that amplify all 13 CODIS STR loci, such as AmpFℓSTR® Identifiler® (ABI) and PowerPlex® 16 (Promega) or that make up the core STRs through the use of a combination of kits, for example the AmpFℓSTR® Profiler Plus® and AmpFℓSTR® COfiler® kits (both from ABI). Kit selection depends on the user’s purpose, as well as DNA availability. STR analysis employs fluorescently labeled primers 5. All currently available ABI kits use four fluorescent dyes. The dyes are excited by an argon laser that emits light at 488 and 514.5nm and causes 5-FAM to fluoresce in the blue region, JOE in the green, NED in the yellow, and ROX (used to label the internal standard for sizing DNA) in the red region of the spectrum 5. As far as determining which dye is used to label each locus, the number of STRs amplified per kit as well as the expected range of possible alleles are taken into consideration. For this project, the 
 16
  AmpFℓSTR® Profiler Plus® PCR Kit was used to conduct all STR analysis. It contains primers that amplify for 9 out of 13 CODIS loci (D3S1358, vWA, FGA, D8S1179, D21S11, D18S51, D5S818, D13S317, and D7S820) and also includes the Amelogenin gene for gender determination.  1.6 STR Analysis 
 1.6.1 Capillary Electrophoresis CE is a separation technique that relies on the application of an electric field to move molecules through a medium confined within a narrow tube. A major advantage to CE is that many systems are fully automated for injection, separation, detection and visualization allowing for many samples to be processed with little or no intervention, for example the ABI Prism 310 Genetic Analyzer. Another advantage of CE is that it can withstand high voltages since separation occurs inside a very narrow capillary, which has a very high surface area to volume ratio. Heat is able to escape the system at a much higher rate, thus allowing the system to be exposed to voltages of up to 800V/cm 18. Data is available within minutes rather than hours of a sample coming off a run. Since the ABI Prism 310 Genetic Analyzer is a popular system that is used by a vast majority of forensic laboratories and was used to collect data for this project details outlined here will be in reference to this instrument. Prior to the injection of each sample, fresh polymer is pumped into the capillary while the preexisting polymer is flushed out. An electric field draws the sample up into the capillary for separation and as alleles migrate toward the anode and separate they pass an open window in the capillary. Here, a laser excites the fluorescently labeled STR products, emission is detected, and information related to fragment size based on distance traveled in the capillary from the time of entry is converted into an electronic file and stored on the computer for further analysis.  
 17
  1.6.2 Genotyping After the separation of STR products, DNA fragments are sized and assigned genotypes. This process is automated using collection and analysis software like GeneScan® and GeneMapper ID®, respectively. Results are displayed on an electropherogram where alleles appear as peaks; this is the DNA profile. As in slab electrophoresis, the size of STR products separated by CE is determined by referencing the travel time of the bands to a ladder run alongside the samples. Allelic ladders are specifically designed for each commercial STR kit and contain all common alleles for each locus examined. In cases where a rare allele appears in a locus and is not found within the allelic ladder, referred to as a microvariant or an off ladder allele, the STRbase is a point of reference containing a list of rare alleles that have been found since the development of the kit. Additionally, an internal sizing standard added to each sample is useful for determining the precision of CE since its peak sizes and intensities should be consistently uniform 19. Miscalled peaks or non-uniformity often indicate a problem with the sample, injection or run.  1.6.3 Interpretation Although allele calls are automatically processed and assigned by a computer, it is the DNA analyst’s responsibility to evaluate the accuracy and quality of the profile. To standardize and ensure consistency of the interpretation of STR profiles across forensic laboratories, the Scientific Working Group on DNA Analytical Methods, sponsored by the FBI, published a document for interpreting autosomal STRs. This guideline describes thresholds for allele detection, identification and interpretation, addresses the interpretation of mixture samples, and demonstrates appropriate statistical calculations 20,21  . Any time DNA evidence is used in court there is much at stake should the DNA  profile be incorrectly analyzed or interpreted. Additionally, most forensic exhibits  
 18
  obtained from crime scenes are too small or so precious that resampling to perform a second analysis is not an option. For these reasons, a host of laboratory controls are necessary and examined before the profile in question can be reported. Controls that are always run alongside an unknown source of DNA include positive and negative extraction and PCR controls. Positive controls of a known source of DNA allow the analyst to conclude that the method of extraction and PCR amplification were conducted properly since the expected result from the positive control is obtained. Negative controls (no DNA included) ensure that contamination from other sources of DNA was not an issue. Higher confidence that the obtained DNA profile is the true profile belonging to the person in question is achieved if these controls give expected results.  1.7 Thesis Objectives Common items that are known to carry PCR inhibitors through extraction include bones, teeth, bloodstained concrete, soil, and denim 22,23. In cases involving these items, the potential problems not only come directly from the substrate itself—for example, calcium from bone, humic acids from soil, and indigo dye from denim—but from the environment in which they are found. Typical insults include variable humidity, UV exposure, precipitation, and bacterial and fungal growth 1. These factors make obtaining a DNA profile even more challenging since much of the DNA will have been degraded. The development of an extraction technique that is not only efficient in eliminating inhibitors but does so without compromising the yield of intact DNA is most desirable. There are a number of ways in which PCR can be inhibited. Inhibitors can bind to either single-stranded or double-stranded DNA, which would make them likely to copurify with DNA and subsequently block amplification; they can interact with the polymerase and repress its enzymatic activity or prevent it from binding to DNA; or they could constrain one of the other components of PCR like magnesium or free nucleotides 22  . Some known inhibitors of PCR are calcium, humic acids, tannins and other  polyphenols, indigo dye (from denim), blood components like heme and hemoglobin, melanin, and collagen 22,24. Phenol, ionic detergents like SDS, and excess salt from the  
 19
  lysate that carry through during the extraction process can also inhibit PCR 22. While samples are often diluted to increase the exposure of PCR reagents to DNA and decrease inhibitors, this approach is not suitable for trace amounts of DNA. Removing inhibitors during DNA extraction and purification is critical to obtaining high-quality profiles from degraded samples; this is the focus of the SCODA technology. In this thesis, the author presents two advantages of using SCODA on difficult forensic samples. The first demonstration employs SCODA as an end-step purification method on bone samples that have previously failed using phenol-chloroform extraction. The second demonstration shows how sample preparation protocols can be optimized for SCODA, in this case to extract DNA from bloodstains on concrete. Finally, to ensure that SCODA can be applicable in real-word scenarios, two actual forensic exhibits are examined. SCODA is shown to achieve higher DNA yield with less inhibition than other extraction methods, leading to the obtainment of STR profiles from otherwise inhibited samples that cannot be used for DNA analysis.  
 20
  Chapter 2 – SCODA 2.1 SCODA Principle SCODA is a nucleic acid concentration and purification method that relies on the electrophoretic motion of molecules in a gel. Unlike slab gel and CE methods where the goal is to achieve size separation, SCODA employs a novel physical parameter to specifically concentrate all nucleic acids from a sample into a single focus location. The parameter k is associated with the nonlinear motion of long, charged molecules in a gel in response to electric fields 25. The fundamental concept of SCODA is to generate periodic motion of DNA using electric fields while synchronously altering the mobility or drag coefficient of the molecule with a second electric field to cause a net drift and its subsequent concentration 25. Successful concentration relies on molecules with large k, which is unique to long polymers such as DNA. This allows SCODA to efficiently select only nucleic acids for concentration, separating them from other molecules in a sample including PCR inhibitors. Additional details regarding the physics behind SCODA can be found in J. Pel et al., 2009. SCODA is applied using a cartridge containing six buffer chambers, their associated electrodes, and one sample chamber (Figure 2.1). The extraction and purification process is broken down into three main processes, which will be discussed in further detail in the following sections. Briefly, they include the injection of a sample into the concentration gel, a wash step that selectively retains molecules with large k in the concentration gel, and a final focus step that concentrates DNA in the gel’s central extraction well. Samples are loaded into the sample chamber, which holds up to 5ml of sample. During injection, an electric field is applied across the chamber and gel to transfer DNA from the sample chamber into the concentration gel while excluding insoluble debris and molecules without a negative charge. Once the sample chamber is depleted, the SCODA electrophoretic fields are applied to begin the concentration process. The wash consists of applying synchronously rotating electric fields with a bias  
 21
  towards one end of the gel to eject molecules with small k. This step is necessary to ensure that negatively charged contaminants brought into the gel during injection are not present in the focus location. The final focus process is identical except for the omission of the bias. At the end of a run, DNA that has focused within the buffer-filled extraction well can be collected and used immediately for downstream processes.  Figure 2.1 – Top view of a disposable cartridge. Buffer chambers (A-F) and their corresponding electrodes (black dots). Image reprinted with permission from Boreal Genomics Aurora User Manual, 2010.  
 22
  2.2 Process  2.2.1 Qt Injection The first process of SCODA is electrokinetic injection, where an electric field is applied across the sample chamber and gel to transfer negatively charged molecules from the sample into the gel. Injection duration is measured by the total electric charge run through the sample and gel 26. Each gel type, depending on the strength of buffer it is made from, has a fixed charge it can accept before any short DNA molecules with high mobility will overrun the gel. Injection stops once the charge threshold, Qt, is reached. Because this charge-based injection is resilient to variations in sample conductivity, applied field, temperature, and other factors, it is more reproducible than specifying a fixed injection time 26. Figure 2.2 is a time-lapse sequence demonstrating injection and concentration.  Figure 2.2 – SCODA injection and concentration. Images A-D is a time-lapse sequence demonstrating injection and concentration of 200ng of SYBR® Green I stained pUC19 DNA. DNA is injected from 5ml of 0.05xTBE buffer into a 1% agarose gel made with 0.25xTBE buffer. Image A is taken after 10min of injection at 20V/cm, B is taken after 10min of subsequent SCODA concentration, and C-D are taken at incremental 20min concentration intervals for a total run time of 60min. Camera exposure times: A=1 000, B=500, C=100, D=20ms. Image reprinted with permission from J. Pel et al., 2009.  
 23
  2.2.2 Washing and SCODA Focusing During injection, any negatively charged molecules in the sample will enter the gel. When the SCODA focusing fields are applied, only nucleic acids with a large k value will focus in the centre while other negatively charged molecules, which have negligible k, move in circles in response to the rotating electric fields 25. Essentially, these other molecules have no net motion and remain in the gel. To ensure that a pure DNA sample is obtained, a slight bias is multiplexed with the rotating fields. The bias allows for molecules and potential inhibitors to be rejected from the gel but still permits DNA focusing; this process is referred to as electrophoretic washing (Figure 2.3). Although the chosen bias is typically not strong enough for DNA to be washed from the gel, it does cause the focus spot to shift from the centre towards the bias. To compensate for this shift, the last process involves applying the same SCODA focusing fields without the bias, which brings the focus point back to the centre of the gel.  Figure 2.3 – Electrophoretic washing of inhibitors. A – D is of a time-lapse experiment in which a mixture of 60µg/ml humic solution and 200ng puc19 DNA were injected, washed and focused in a SCODA gel. The increase in clarity of the gel is indicative of the humics being expunged from the gel. D. UV‐transilluminated image (100ms exposure) taken at the same time point as image C (80min total elapsed time), in which stained DNA is clearly visible in the centre of the gel. Image reprinted with permission from J. Pel et al., 2009.  
 24
  An advantage of SCODA is that it can concentrate DNA up to 100-fold because it can take up to 5ml of sample and focus it into a volume of 50µl with or without a wash step. When the wash step is added, DNA quality is further improved by removing inhibitors. One disadvantage is that because concentration is dependent on k, shorter DNA molecules (< 300bp) with small k will not focus. This limitation would not affect its application to degraded samples. DNA degradation, which is the fragmentation of the whole molecule, is completely random. The likelihood that a DNA fragment smaller than 300bp will contain a targeted STR locus is extremely small. SCODA can concentrate partially degraded DNA containing fragments greater than 300bp from which the targeted STR loci can be identified.
 
 2.3 Instruments Two SCODA instruments built by Boreal Genomics Inc. were used in this project: the HsBB and the Aurora pictured in Figure 2.4. Aside from their aesthetic differences, these instruments contain: an electrode plate, power supplies, a cooling system, and a computer to control the processes. Additionally, both are equipped with an imaging system, which has the capacity to automatically take and collect images during a run if the DNA sample is stained with a fluorescent dye, like SYBR® Green. Essentially, the Aurora is a newer, more compact version of the HsBB with the main difference being the type of cartridge used. At the time of its development, disposable cartridges were not made available for the HsBB and reusable gelboats had to be manually washed and sterilized using bleach and UV-crosslinking. Further, agarose gels had to be made fresh each time an experiment was conducted. Seeing the need for consistency, disposable cartridges containing precast gels, running buffer and electrodes were designed for the Aurora to reduce variability stemming from self-prepared gels and running buffer. The HsBB platform has been discontinued since the Aurora’s introduction. As with all electrophoresis systems that use electric fields to drive molecules, a large amount of heat is generated throughout the duration of a run. For this reason, slab gels are limited by the amount of voltage applied. CE is confined to a narrow tube that  
 25
  exchanges heat easily but the Aurora has a cooling system to remove heat from the system to maintain adequate performance and prevent the gel from melting 26.  Figure 2.4 – Prototype SCODA instruments. The Aurora (left) and HsBB (right) SCODA intruments are pictured with cartridges that are sat on top of a cold plate connected to a water chiller (not shown). The processes of both instruments controlled by a computer. Image of the Aurora reprinted with permission from Boreal Genomics Aurora User Manual, 2010.  
 26
  Chapter 3 – Bone Samples 3.1 Introduction Bones and teeth are the best sources of hard-tissue DNA for forensic investigations but analysis is often limited by a number of factors including limited amounts of available endogenous DNA, significant contamination from exogenous DNA, degradation, and high concentrations of PCR inhibitors 13,23,27,28. Many extraction and purification methods have been designed to obtain as much DNA as possible while minimizing sample handling and the amount of co-extracted inhibitors. Despite this, a small percentage of samples still fail to yield a profile 13. In these cases, it is difficult to assess whether an inhibited sample has enough DNA to begin with or if the DNA is just being masked by the presence of inhibitors. One of the primary inhibitors found in many bone cases is humic acid, a class of chemical compounds that are abundant in soil and water 29. DNA extracted from skeletal remains that have been buried and exposed to soil for various periods of time experience PCR inhibition from humic acids and give false negative results. Humic acids bind to Taq polymerase, which causes a decline in PCR efficiency and ultimately DNA yield 30. Phenolic groups in humic acids may also become oxidized and covalently bind to DNA 29,31  . Another complication in extracting DNA from bone is in its physical nature. Bone  is a connective tissue that mainly consists of the protein collagen and inorganic mineral hydroxyapatite (70%), which includes calcium phosphate, calcium carbonate, calcium fluoride, calcium hydroxide and citrate 27. Calcium from hydroxyapatite has a strong affinity to the phosphate backbone of DNA and creates a physical barrier between lysis reagents and cells, which prevents DNA from being released. However in 2007, Loreille et al. found that EDTA when used at a high concentration (0.5M) and volume (10ml) to dissolve 1g of bone powder, it can completely demineralize hydroxyapatite and improve  
 27
  DNA yields. EDTA is a strong chelator of bivalent metal cations like Mg2+ and Ca2+, which are required for the function of many DNAases that degrade DNA. For this reason, it is often a component in most lysis buffer recipes, albeit at a much lower concentration, to reduce undesired degradation. Phenol-chloroform coupled with centrifugal filtration concentration is the gold standard in forensic science for DNA extraction from large sample volumes, including bones. The method is used by the RCMP and performs well on fresh bone samples but poorly on ancient bones or severely degraded bones exposed to harsh environments 32. Because PC extraction relies on differential solubility of DNA and other molecules, inhibitors from samples with similar solubility properties to DNA like humic acids are co-extracted. This becomes most problematic during PCR since humic acids and other inhibitors that carry through extraction inhibit the reaction and samples must be diluted until the sample is favourable for PCR 13,33. However, diluting degraded or low quality DNA could also result in allelic dropout, low signal strength or non- and partial profiles due to stochastic sampling 34. Another strategy used to remove PCR inhibitors from DNA is to perform a second extraction, which may or may not reduce inhibitor levels. This difficulty in removing humic acids from bone samples makes SCODA a good candidate for DNA purification. In this study, nine bone samples (6 femoral, 1 mandibular, 1 vertebral, and 1 cranial) were obtained from the BCCS and subjected to PC extraction or PC extraction followed by SCODA purification. DNA extraction results from two femoral bone samples (BB7 and BB9), which feature the use of SCODA are presented in detail below as examples. EDTA lysis was also attempted for BB9, which contained very little DNA. The bones that were previously processed by another laboratory for the BCCS using PC extraction did not yield an STR profile so the goal was to determine if the failure was due to PCR inhibition and if a profile could be obtained after SCODA purification. DNA recovery was determined by the Quantifiler™ Human DNA Quantification Kit and inhibition was assessed by comparing IPC CT values of samples to those of the standard curve run on the same reaction plate.  
 28
  3.2 Methods and Materials Sample Preparation DNA extraction, pre-PCR set up and amplification processes were all carried out in separate designated rooms to minimize the risk of cross contamination. BB7 and BB9 were also extracted at different times; this follows the reduction in potential DNA contamination protocol common in forensic science laboratories that is referred to as “separation of samples in time and space”. Femurs were cleaned with 10% bleach followed by 99% ethanol. A reciprocating scroll saw (Dremel benchtop model) was used to cut a sagittal section of the bones (Figure 3.1), which were then cryogenically ground into fine powder in a freezer mill (SPEX Corporation, model 6750) with liquid nitrogen. These procedures were in compliance with the Standard Operating Procedures at the BOLD Laboratory.  Figure 3.1 – Cut Sagittal-section of a femur: sample BB7.  
 29
  Sample Lysis and Extraction Bone powder was divided into four 15ml centrifuge tubes each containing 1.0g. All four replicates, a Blood Internal Standard (BIS) positive control of known DNA, and a negative control were each placed in 3ml of lysis buffer (10mM Tris-HCl pH=8.0, 10mM EDTA pH=8.0, 50mM NaCl, 2% SDS, and 1.5mg/ml Proteinase K) and incubated at 56°C overnight. Initially, BB9 was lysed in a similar manner as BB7 but after determining there was insufficient DNA in 1.0g of bone by qPCR for STR amplification a larger piece yielding 46g of powder was processed. The powder was subjected to EDTA lysis in demineralization buffer (0.5M EDTA, 1% SDS, and 1.5mg/ml Proteinase K) to maximize the amount of DNA that could be obtained. To maintain the 1:10 ratio of bone powder to lysis buffer required for complete demineralization 27, 50ml was applied to each 5g powder and incubated at 56°C overnight prior to being centrifuged at 10 000g for 20min to pellet any bone powder not dissolved. The supernatants were then combined, diluted by 10-fold and concentrated in 12x Amicon Ultra-15, 30kDa centrifugal filter columns (Millipore) until only 200µl was left in each filter. The residual concentrates were then pooled and subjected to PC extraction. Equal volumes of phenol:chloroform:isoamyl alcohol (25:24:1) were directly added to the lysates of all replicates and centrifuged at 1 000g for 5min. Supernatant from each sample was carefully transferred to a new tube and the process was repeated. A final extraction step with N-butanol was performed to ensure the removal of excess phenol carry-through. The supernatant was then concentrated in a Centricon YM-100 (Millipore) and washed twice with TE buffer at 2 000g for 30min before being eluted in 50µl of the same buffer. SCODA Purification Fifty microliter outputs obtained from PC extraction were diluted to 5ml with FAD H2O in the SCODA sample chamber. BB7 replicates and controls were processed on the HsBB instrument since the Aurora was not available at the time of testing, while  
 30
  BB9 and controls were carried out on the Aurora instruments with disposable cartridges. Details pertaining to gel and run conditions can be found in Appendix A. Quantification, PCR inhibition and STR Profiling Quantification and STR profiling of DNA were performed using the Applied Biosystems Quantifiler™ Human DNA Quantification Kit and the AmpFlSTR® Profiler Plus® PCR Kit, respectively. Twenty-five microlitre reactions were used for both assays with 2µl and10µl input template DNA for the Quantifiler™ and the Profiler Plus® kit, respectively. Amplification was carried out according to manufacturer specifications. PCR inhibition was assessed by comparing IPC CT values in each sample well to the IPC CT values of the standards from the same reaction plate, which was typically averaged to be ~28 cycles. Therefore, a sample with an IPC CT delay of 28 cycles is considered completely inhibited. STR products were separated and visualized with the ABI Prism™ 310 Genetic Analyzer. Electrokinetic injection and electrophoresis run times were set to 5s and 24min respectively at 10kV. ABI Prism 310 Genetic Analyzer Data Collection Software v3.1 and GeneMapper ID v3.2 were used for data collection and analysis. The threshold range for an allele call was 50 to 4 000RFUs.  3.3 Results Nine bone samples of various ages and stages of decomposition were submitted by the BCCS for the study. These samples had previously been processed before using PC extraction but only non-profiles or partial profiles were obtained. In these samples that were subjected to DNA profiling, three possible outcomes were anticipated: 1) PC extraction with or without having to dilute inhibitors was sufficient to obtain a profile indicating that there was an adequate amount of DNA in the sample,  
 31
  2) The sample is so overwhelmed with inhibition after PC extraction that even dilution masks the presence of DNA giving a false negative, providing a case for the use of SCODA for purification, 3) The sample failed completely—no profile can be obtained with PC and/or with SCODA suggesting a lack of sufficient DNA. Table 3.1 lists eight out of the nine bone samples that were assayed. Of these samples, 62% failed completely with PC extraction and with or without SCODA, 38% gave full profiles by extracting with PC only and no samples provided a case for SCODA purification. However, of the samples that gave full profiles, amplification of BB7 was possible only after being diluted from the PC extract; this same sample also gave a partial profile (8/9 loci) after being purified with SCODA (Figure 3.2). To mimic an ideal scenario for the use of SCODA, BB9 was processed with PC extraction at a much larger scale (~46x); amplification during quantification was possible only after being subjected to SCODA purification, but the sample still gave no profile (Figure 3.3).  
 32
  Sample Name  Type of Bone  Profile (F/P/N)  BB1  Femur  F  BB2  Femur  F  BB3  Cranial  N  BB5  Mandible  N  BB6  Femur  N  BB7  Femur  F  BB8  Femur  N  BB9  Femur  N  Table 3.1 – Summary of processed bones from the BCCS. Results shown are those as extracted by PC; full (F), partial (P), or non-profiles (N) were expected. Samples that did not yield profiles were subsequently purified with SCODA. In Figure 3.2a, the total DNA recovered from positive and negative controls as extracted by either PC or PC followed by SCODA are comparable. For BB7, undiluted PC extracts are completely inhibited—amplifying more than 25 cycles after the standards—while those purified with SCODA are not. However, after diluting samples at a ratio of 1:5 DNA yields from PC are greater than SCODA, which amplified below the standard curve at 1:5. Further, only a partial profile could be obtained from the undiluted SCODA extracted compared to a full profile obtained by amplifying the 1:5 diluted PC extract (Figure 3.3).  
 33
  Total DNA (ng)  a. DNA Recovery - Quantifiler 140.00 120.00 100.00 80.00 60.00 40.00 20.00 0.00  1:1 1:5 1  2  1  PC  2  PC  SCODA  PC  SCODA  SCODA BB7-1  BIS11  Ext Neg  IPC Delay (# of Cycles)  b. PCR Inhibition - IPC 30.00 25.00 20.00 15.00 10.00 5.00 0.00  1:1 1  2  1  PC  2  PC  SCODA  PC  SCODA  1:5  SCODA BB7-1  BIS11  Ext Neg  Figure 3.2 – Quantifiler™ Human DNA Quantification Results for BB7. (a) Quantifiler™ Human DNA quantities of BB7-1 as extracted by PC or PC followed by SCODA; two replicates each. Undiluted samples are denoted as 1:1, while 1:5 indicates that the sample was diluted by a factor of 5. The beige bar indicates that DNA amplified below the standard curve. (b) PCR inhibition represented as the number of IPC threshold cycle (CT) delays from averaged IPC CT values of DNA standards from the same run.  
 34
  Figure 3.3 – STR profile of BB7 as purified by SCODA.  To determine if the larger alleles could be detected and visualized, the amplified sample was reinjected at an extended time of 10s during CE. Figure 3.4a is the full STR profile that was obtained by just increasing injection time. Figure 3.4b is enlarged to show the overlap of dyes in the region where additional peaks appeared.  
 35
  Figure 3.4 – STR profile of BB7 purified by SCODA with increased injection time. (a) A full profile is obtained after increasing the electrokinetic injection time from 5s to 10s. Two extraneous peaks are pulled up as a result of the increase in signal strength and saturation of the matrix. (b) Note that the green peaks (pull-ups) are directly underneath the larger blue peaks. 
 36
  Forty-six grams of BB9 bone powder were processed and eluted into the same 50µl volume as a standard extraction from 1.0g of bone powder. The idea was to increase total DNA yield and overwhelm the sample with inhibitors to the extent where diluting would not only remove inhibitors but also reduce the amount of DNA below detectable limits. Dilutions at 1:1, 1:5, and 1:10 of the PC output were quantified. Figure 3.5 shows that PC extracts were completely inhibited at 1:1, 1:5, and 1:10 dilutions whereas SCODA samples were not inhibited in the undiluted extract and could not be amplified in diluted samples. The PC extract was also re-quantified at 1:100 dilution, which showed no inhibition according to the IPC but, was too dilute for quantification (data not shown). The positive and negative controls are shown in Figure 3.6. Surprisingly, although Quantifiler™ indicated sufficient DNA was present in the sample for STR amplification, no profile could be obtained; the positive control ran through SCODA also failed to yield a profile.  
 37
  Total DNA (ng)  a. DNA Recovery - Quantifiler 10 9 8 7 6 5 4 3 2 1 0  1:1 1:5 1:10 PC  SCODA-spVac BB9  IPC Delay (# of Cycles)  b. PCR Inhibition - IPC 30.00 25.00 20.00 15.00 10.00 5.00 0.00  1:1 1:5 1:10 PC  SCODA-spVac BB9  Figure 3.5 – Quantifiler™ Human DNA Quantification Results for BB9. (a) Quantifiler™ Human DNA quantities of BB9 as extracted by PC or PC followed by SCODA (SCODA). Undiluted, and diluted samples at 1:5 and 1:10 were assayed. Beige bars indicate amplification was below standard curve (b) PCR inhibition represented as the number of IPC CT delays from averaged IPC CT values of DNA standards from the same run.  
 38
  Total DNA (ng)  Quantifiler - Extraction Contols 5.00 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00  *40 ng  1:1 1:100  Pos  Neg PC  Pos  Neg  SCODA-spVac  Figure 3.6 – Quantifiler™ results for BB9 extraction controls. Positive (Pos) and Negative (Neg) extraction controls were processed alongside BB9 to ensure the extraction technique was successful and that no contamination occurred, respectively. The asterisk (*) indicates a quantity above the maximum plotted DNA yield.  3.4 Discussion Hydroxyapatite and collagen are the main components of bone and play a significant role in preserving DNA from degradation. The cells containing DNA are embedded within the composite of the bone matrix. This makes bone a good source of DNA but extracting it from these samples remains a challenge, especially when inhibitors like humic acids are present. In this study, several bone samples were subjected to PC extraction and some were further purified with SCODA depending on the abundance of DNA and severity of inhibition. The aim was to identify when inhibition was more problematic than the inadequacy of DNA and determine if SCODA could be a plausible solution to obtaining a profile.  
 39
  Successful SCODA concentration relies on the nonlinear velocity behavior of DNA in a gel matrix in response to an electric field 25,14. This interaction is strong with long DNA, but DNA that is shorter than 300bp cannot be focused 25. In non-degraded samples, DNA yields should appear comparable between PC extraction and PC followed by SCODA extraction. In samples that are more degraded, quantification may be biased against SCODA since the amplification target of Quantifiler™ is only 62bp, which can still detect fragments shorter than 300bp in PC extracts. Figure 3.6 is an image of a slab gel comparing BB7 as extracted by PC without and with SCODA where the higher SCODA length cutoff can be seen. There is a discrepancy between the DNA recoveries of PC extraction and SCODA. On the other hand, if this were true we would expect SCODA to yield a full profile and not have difficulties amplifying larger STR fragments since longer fragments are captured. This means that although larger fragments are captured, they may not all contain the regions necessary for STR amplification.  Figure 3.7 – Slab gel electrophoresis of DNA extracts from BB7. 1kb and 100bp ladders are run in parallel with samples for size reference. Two replicates of DNA from BB7 and a positive control (BIS11) as extracted by PC or PC and SCODA (SCODA) are shown. In BIS11, where DNA is not degraded, fluorescence intensity and length recovery is comparable between PC and SCODA while in BB7, where DNA is degraded, a cut off at ~300bp in the SCODA outputs can be seen. 
 40
  Applying SCODA to a DNA extract that came from another extraction method reduces its yield by at least 30%. The SCODA extracted sample of BB7 gave a partial profile due to stochastic sampling of an already reduced amount of total DNA. The sample would contain fewer larger STR regions for amplification and be below the detection threshold (50RFU). This phenomenon is also commonly seen in degraded samples, where preferential amplification for smaller fragments occurs 35-37. For SCODA to be successful on forensic exhibits, the initial sample must not be too degraded and have enough STR amplifiable DNA that when first processed by a method like PC extraction can still take a 30% reduction and give a full profile. A longer injection or injection at a higher voltage during CE allows for more products to enter the capillary to improve the detection of alleles in partial STR profiles from low-level DNA samples 38. A downfall of this process however, is the potential to pull up non-allelic peaks, which could further complicate interpretation. Spectral pull-ups are a common example of non-allelic peaks that occur when a saturated signal from one dye channel is mistakenly interpreted as being from another channel. Therefore, the pullups observed in BB7 can be eliminated since the cause of the spectral pull-ups can be attributed to an overall increase in the signal strength and saturation of the matrix, or background. PCR inhibition, DNA degradation and low yields are present in accordance with each other. A heavily degraded sample is overwhelmed with inhibition and low yields and vice versa but to find a heavily inhibited sample that also contains high concentrations of good quality DNA is rare. Therefore, given the small sample size, it was not a surprise that a bone sample where SCODA would be applicable was not available. Does such a sample even exist? In a real-world scenario if there was a possibility that a person could be identified by using more starting material regardless of processing time, this would be conducted. Therefore, to investigate if increasing the amount of starting bone powder would also increase the amount of total DNA to yield an STR profile, BB9 was revisited and ~46g of it was processed; a positive and negative control was processed similarly. However, one caveat to using more material is that this  
 41
  also increases the amount of inhibitors. This was evident in the quantification results where the PC extract was completely inhibited and could not be quantified even in the diluted samples since inhibitors were probably saturating the sample due to increasing the starting material. When SCODA was applied, amplification was possible in the undiluted extract where the IPC were indicative of being inhibitor-free and the yields were sufficient for STR amplification also. It was surprising then that no profile could be obtained from this sample or the positive control despite the encouraging quantification results of both. Several factors could have prevented the STR amplification of BB9. The method that was used for processing 46g of BB9 bone where the supernatant was concentrated prior to carrying out PC extraction is not a common procedure but was conducted to avoid performing extractions on > 1L volumes with phenol. It is important to keep detergents in solutions being concentrated by centrifugal filters below their critical micelle concentration (CMC). Detergents that are above their CMC will form lipid structures called micelles that could envelop DNA 39. That is why the supernatant was diluted by a factor of 10 to prevent SDS from reaching its CMC of 7mM 40. However, successive concentration on the centrifugal filter could have increased the concentration of SDS. If micelle-DNA complexes begin to form because SDS reaches its CMC, the aggregates may become too big to flow through the membrane and end up in the DNA output, which could inhibit PCR 22. Quantification PCR may not have been affected completely if only some DNA was bound to SDS; its ability to be amplified would be reduced and so would its yield but the STR assay is much more sensitive to inhibition since more primers and larger amplification regions are involved. Additionally, because the IPC is not exposed to the same SDS-bound inhibition, its amplification may not be affected either. The same argument can be made for the positive control, which also contained sufficient DNA but did not yield a profile. Sutlovic et al. (2008) suggests that humic acids bind to Taq polymerase and at elevated concentrations can also covalently bind to human DNA making it unrecognizable for Taq polymerase. It is possible that although SCODA was able to  
 42
  remove a substantial amount of humic acids, some might have directly bound to the DNA affecting STR amplification. This inhibition would be more detrimental to amplification with Profiler Plus® than with Quantifiler™ since more primers and larger template regions ranging from 100-350bp are required in the former than compared to 62bp in the latter as previously mentioned. This ultimately causes multiplex reactions to be more sensitive to inhibition since more interactions are present in a single reaction and longer amplicon sizes require higher processivity from Taq polymerase 41. Further, this mechanism of inhibition could not have been detected by the IPC since the control template was not exposed to the same humic environment as the sample DNA. To rule out the possibility that SCODA may produce a by-product that could inhibit STR amplification, buffer from a negative control run on SCODA was used to poison the standard 9947A. Amplification was carried out on several dilutions of the DNA standard using the negative control instead of water; as little as 100pg of the standard was analyzed. DNA profiles were obtained from all dilutions and no obvious deficiencies or anomalies were apparent (no data shown). This means that SCODA and the buffer from the extraction well does not affect or inhibit STR analysis. Despite not having the ideal sample for using SCODA within our sample set, SCODA has demonstrated the ability to remove inhibition and allow for STR amplification resulting in DNA profiles from undiluted extracts. Further, though a profile was not obtained in BB9, the sample made to resemble our ideal scenario where inhibition and total DNA is high, DNA was still quantifiable post SCODA purification over the completely inhibited PC extraction. This shows that SCODA can improve inhibition by at least 10-fold and perhaps obtaining an STR profile would be possible by optimizing the upfront extraction method to avoid the formation of inhibitor-DNA complexes.  
 43
  Chapter 4 – Bloodstains on Concrete 4.1 Introduction Concrete is a composite building material that contains high concentrations of calcium and silica, two elements that are known to interfere with the process of extracting DNA for analysis from biological stains on concrete 32,42. The exact mechanism is not known but stains on concrete have proved to be challenging especially in cases where only trace amounts of biological material are present 32. For extraction methods that rely on binding DNA to silica columns or resins, it is thought that the silica from concrete competitively binds to DNA 32. Generally, stains are collected from a crime scene using cotton swabs. In 1997, Sweet et.al. showed that the double swab technique improves DNA yields for collecting saliva from bitemarks. The technique uses a wet swab followed by a dry swab to rehydrate and collect cells from the stain, respectively. Since then, it has been adapted for the collection of stains from various substrates. However, because concrete is a porous matrix that allows for deposited biological fluids to permeate, swabbing with any number of swabs may fail to collect any material trapped beneath the surface. Further, DNA loss is significant during extraction from a swab since cells can become trapped within the fibres of the cotton swab 39,43. Therefore, it would be ideal to collect a stain in its entirety and directly apply lysis buffer to extract as much DNA as possible. At the very least, it is desirable to collect a representative sample of the stain if it cannot be collected in its entirety. The compatibility of using a lysis method that contains high concentrations of EDTA for bloodstains on concrete (BSC) was investigated for its use in conjunction with SCODA. The thought is that high concentrations of EDTA would counteract high concentrations of calcium present in concrete. Promega’s silica-based, paramagnetic bead DNA IQ™ system, used by the RCMP to process small volume casework samples  
 44
  including swabs, blood cards, and swatches of fabric 32, will be used in addition to PC extraction as a control to benchmark the SCODA protocol. Chipping sections of the concrete with representative stains was also examined as an alternative to swabbing for the collection of DNA from stains on concrete to improve the recovery of usable DNA. To compare whether chipping or swabbing yields more DNA, a protocol was designed and optimized for chipped bloodstains so that both collection techniques could be compared appropriately. Chipping a stain should allow the collection of more cells since bloodstains penetrate the surface of the concrete, but at the risk of associating the sample with increased concrete particulates during lysis (Figure 4.1a). A method that is more robust to high mineral levels is necessary to recover DNA from chipped samples.  4.2 Methods and Materials Sample Preparation and Collection A 12x12 grey paving stone (Home Depot) was divided and marked into one-inch squares and 10µl or 1µl of blood from a known donor was applied to the centre of each square. Positive and negative controls for each extraction method were prepared with 10µl or 1µl of blood or FAD H2O in 1.7ml microfuge tubes. After allowing stains to dry overnight, replicate samples were collected either using the double swab technique 44 or chipping. A hammer and Robertson screwdriver were used as a mallet and chisel to chip the stains from concrete (Figure 4.1b). A folded piece of paper towel was placed above the stain to contain the resultant dust, which was subsequently collected in a microfuge tube with forceps (Figure 4.1c). To minimize sample-to-sample collection variability, chipped BSC from each square was combined within the 10µl or 1µl samples prior to being  
 45
  evenly distributed by mass for lysis while those collected by swabbing were pooled after being lysed.  Figure 4.1 – Collection of bloodstains on concrete. (a) Chipped stains are shown next to swabbed stains. Residual blood can be seen from the surfaces where the stains were swabbed. (b) A pair of forceps, a Robertson screwdriver and a hammer was used to collect and chip BSC. (c) A folded piece of paper was used to contain the chipped fragments. Circled in red is a 1µl BSC.  
 46
  Sample Lysis Samples were lysed in standard buffer (10mM Tris-HCl pH=8.0, 10mM EDTA pH=8.0, 50mM NaCl, and 2% SDS) or EDTA buffer (0.5M EDTA, 1% SDS) with 1.5mg/ml Proteinase K, while Promega samples were soaked in 50µl FAD H2O and incubated at 56°C overnight. Post incubation, samples were centrifuged at 13 000rpm for 5min to collect supernatant. Promega Extraction Due to the increased concentration of silica in chipped concrete stains, it was recommended by Promega that the BSC be soaked in FAD H2O and extraction be performed on the supernatant with the anticipation that cells would diffuse into the water and avoid direct extraction from the concrete. The Promega DNA IQ™ System (DC6701) for small volume casework samples was then used according to the manufacturer’s protocol (part no. TB296) for stains on solid material (4B) and liquid samples (4C) for chipped stains and swabs, respectively. SCODA Extraction For each replicate, the supernatant was diluted by a factor of 10 with FAD H2O and concentrated to 100µl in a Vivacon-2 100KDa filter at 2 000g (Vivaproducts, VN02H21ETO). The retentate was then made up to 5ml with FAD H2O, applied to the sample chamber of a cartridge, and run on the Aurora using the Aurora DNA Clean-up Protocol. DNA Analysis Quantification, PCR inhibition, STR amplification and analysis of DNA were all carried out as previously described in Chapter 3.2.  
 47
  4.3 Results Standard and EDTA Lysis Standard and EDTA lysis buffers coupled with PC extraction and SCODA were compared against each other to determine the best combination of lysis and extraction of DNA from chipped BSC. Diluted PC extracts at 1:10 were assayed in addition to the undiluted samples since inhibition was expected, whereas only undiluted SCODA extracts were assayed; results in Figure 4.2 are consistent with these assumptions. Because the standard lysis coupled with PC extraction is the standard protocol, this will be referred to as just PC extraction. DNA recovery from PC extraction is not consistent between samples but is much less inhibited than EDTA-PC; even the EDTA-PC positive control had to be diluted by a factor of 100 before amplification was possible. EDTASCODA had comparable DNA yields to one of the replicates from PC extraction but results were more consistent between samples and no inhibition was seen in undiluted samples.  
 48
  IPC Deviation (# of Cycles)  b. PCR Inhibition - Lysis and Extraction Combinations 30 25 20 15 10  1:1  5  1:10  0  1:100 1  2  Pos  1  Standard lysis  2  Pos  EDTA lysis  1  2  Pos  1  Standard lysis  PC  2  Pos  EDTA lysis  SCODA  Figure 4.2 – Comparison of standard and EDTA lysis with PC or SCODA extraction. (a) Replicate samples (1-2) and positive controls (Pos) were subjected to standard or EDTA lysis followed by PC or SCODA extraction. Only undiluted SCODA samples were quantified. Undiluted samples are denoted as 1:1, where as 1:10 and 1:100 are samples that have been diluted 10 and 100-fold, respectively. The asterisk (*) indicates quantities above maximum plotted DNA concentration. Standard lysis and PC extraction, and EDTA lysis and SCODA extraction (✓) were the combinations that were chosen for furher chipped BSC testing. (b) PCR inhibition represented as the delay in IPC CT averaged values of DNA standards.  SCODA, PC, and Promega Extraction of 10µl BSC Based on results from Figure 4.2, where DNA recoveries and the absence of inhibition were comparable between PC extraction and EDTA-SCODA chipped samples were prepared using these methods and benchmarked against Promega’s DNA IQ™, the  
 49
  RCMP laboratory’s method of choice for processing small volume samples like stains. Two sets of experiments (data Set A and B) were carried out to compare chipped BSC described in Figure 4.3; Set B includes comparisons to swabs, collected by the double swab technique. Looking at chipped stains only, both data sets show that Promega DNA IQ™ did not yield any quantifiable DNA, while PC extraction and EDTA-SCODA did. In general, recoveries were lower in Set B but within each set EDTA-SCODA DNA recoveries were more consistent between replicates than PC extraction. No inhibition was seen in any of the samples (data not shown).  
 50
  Figure 4.3 – DNA extraction from 10µl bloodstains on concrete. Replicates of 10µl BSC including positive (Pos) and negative (Neg) controls were extracted using the Promega DNA IQ™ System, SCODA, and Standard-PC extraction. (a) Data set A, n=4. (b) Data set B, n=5.  
 51
  Figure 4.4 is a summary showing the mean DNA recovery and standard error by combining replicates from both data sets. DNA yields from Set A were normalized to Set B based on the positive controls, which are consistent between PC extraction and EDTASCODA. Overall, chipping and extraction by EDTA-SCODA is the most effective method for obtaining the highest DNA yield from BSC. The Welch’s t test was performed to confirm that this result is significant (p = 0.01); see Appendix A.  Summary of Swabbed and Chipped 10ul BSC 0.50  DNA Concentration (ng/ul)  0.45 0.40 0.35 0.30 0.25 0.20  Mean  0.15 0.10 0.05 0.00 Standard-PC  DNAIQ  Swabbed BSC  Standard-PC  DNAIQ  EDTA-SCODA  Chipped BSC  Figure 4.4 – Summary of swabbed and chipped 10µl BSC. Replicate samples from data SetA and SetB were normalized based on the positive controls from each set, to give the mean DNA concentration (ng/µl) and standard error (black bars) of swabbed and chipped 10µl BSC samples as extracted by Standard-PC, DNAIQ™, and EDTA-SCODA.  
 52
  Sensitivity Test for Swabbing and Chipping One-microlitre BSC experiments were conducted to determine further whether swabbing or chipping is more efficient for collecting DNA from BSC and to assess concurrently the sensitivity of all methods. Figure 4.5 shows that no amplifiable DNA was recovered from 1µl BSC with DNA IQ™ by using any collection method and within PC extraction, swabbing was also less efficient than chipping. For chipped stains, PC and EDTA-SCODA DNA extraction yields were comparable.  0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00  Swabbed BSC  EDTA-SCODA  DNAIQ  Standard-PC  DNAIQ  Mean Standard-PC  DNA Concentration (ng/ul)  Summary of Swabbed and Chipped 1µl BSC  Chipped BSC  Figure 4.5 – Summary of swabbed and chipped 1µl BSC. Replicates of swabbed (n=5) and chipped (n=8) 1 µl BSC samples were extracted using Standard-PC, DNA IQ™, and EDTA-SCODA extraction. The mean DNA recovery (ng/µl) and standard error (black bars) for each method is shown.  DNA recoveries from both PC extraction and EDTA-SCODA were variable around the required concentration ranges observed for optimal STR amplification (0.08 – 0.125ng/µl). To determine whether DNA profiles could be obtained from such low  
 53
  recoveries, STR amplification and analysis were carried out. Figure 4.6 shows the number of full and partial profiles that were obtained and which alleles were called for each locus. A locus can only be used for interpretation when two heterozygous alleles amplify above the defined threshold of 50RFU (for homozygous, the threshold is 100RFU). More full profiles were obtained by EDTA-SCODA and more partial profiles by Standard-PC but many non-profiles were also observed from both methods. Therefore, it cannot be conclusively determined that one method is better than the other. Figure 4.6b shows that alleles from the largest STR loci: FGA, D18S51, D13S17, and D7S820, were most commonly dropped out.  
 54
  a. Summary of STR Profile Types Obtained from Chipped 1ul BSC 60 50 Percent  40 30  Standard-PC  20  EDTA-SCODA  10 0 Full  Partial  Non-profiles  b. Allele Calls of Chipped 1ul BSC 100  Percent  80 60 Standard-PC  40  EDTA-SCODA 20 0  Figure 4.6 – STR Analysis of Chipped 1µl BSC. (a) Percent of full, partial, non-profiles obtained (n=8). (b) Availability of alleles for interpretation from the 9 loci as amplified by the AmpFlSTR® Profiler Plus® PCR Kit from 1µl BSC extracted with PC or EDTA-SCODA. Loci are in order of appearance on electropherograms.  
 55
  4.4 Discussion DNA recoveries from BSC were consistent with those seen previously in forensic laboratories. Bead-based extractions failed by being exposed to increased concrete concentrations with chipped BSC or was not efficient at extracting DNA from small amounts of blood from swabs. High levels of calcium from concrete causes competitive binding of DNA to silica present in the concrete surfaces instead of the beads upon the addition of binding buffer. PC extractions were observed to be highly variable although chipped samples were homogenized and evenly distributed. Chipping was also shown to be more efficient than swabbing, which was expected since swabbing a porous surface like concrete leaves material behind. DNA and cells can also become trapped within the swab itself, further reducing extraction efficiency. Overall, the best method for obtaining DNA from a minimum of 10µl BSC was chipping in combination with EDTA-SCODA. An important aspect of SCODA injection that affects the development of protocols is that the final sample conductivity must be less than 25% of the gel conductivity; this is to ensure that the DNA from the sample chamber is completely transferred to the concentration gel for an efficient injection. When these conditions are met, DNA mobility through the sample buffer is much more rapid than through the gel otherwise, DNA could migrate off the far end of the gel or remain in the sample chamber and not get focused 25. Therefore, because many of the components in the EDTA lysis buffer is highly conductive precluding the direct use of SCODA, a sample desalting step by ultrafiltration was required. An interesting parallel that concrete has to bone is that they both contain high concentrations of some form of calcium. Immersing bone samples in large volumes of EDTA solution helps to demineralize the hydroxyapatite in bone, allowing lysis buffer access to DNA and preventing DNA degradation by impeding the function of DNAases, as discussed in Chapter 3.1. Therefore, a similar lysis method for bloodstains on concrete (BSC) used in conjunction with SCODA was investigated here and shown to be more  
 56
  efficient at obtaining DNA with SCODA than the standard lysis method. EDTA is used in several applications to disaggregate calcium matrices; for example, an industrial application of EDTA is to dissolve limescale, which is essentially caused by the build up of calcium deposits. The use of EDTA in such high concentrations also has its limitations however, as it is a strong PCR inhibitor. For this reason it is not possible to couple EDTA lysis with PC extraction alone since EDTA is water-soluble and would co-extract with DNA. Although DNA recoveries from EDTA-SCODA was greater than PC extraction as seen for 10µl BSC, yields were quite low compared to their positive controls. There are a few possibilities that could explain this. Lysing material from chipped concrete leaches more calcium into the lysate than lysing swabbed material. This increase in calcium would affect the measurement of DNA yields by competing with magnesium ions, a required cofactor for Taq function, which would reduce the efficiency of the reaction 30. However, co-extraction of Ca2+ is unlikely since there was no effect on the IPC of samples. Another possibility as to why DNA recoveries of samples to positive controls were so low is that the sedimentation of the chipped concrete does not allow for lysis reagents to reach cells efficiently. From the results, the amount of DNA recovered from 10µl BSC was not 10-fold more than the DNA extraction from the early tests with 1µl as expected. This discrepancy can be attributed to the high variability in sample collection as well as the amount of DNA present in each blood sample used to prepare the two experiments. To minimize the variability between chipped samples belonging to the same experimental set, replicates were pooled to homogenize the sample as best as possible prior to being evenly distributed by mass. However, it is difficult to maintain the same mass of concrete to volume of blood ratio. Since the stain covers such a minute area and it is challenging to distinguish the blush coloured stain against the concrete after it has been chipped, it is possible that in some cases too much concrete or not enough blood was collected. Overcollection would unnecessarily increase inhibitor concentration and under-collection could result in recovering low amounts of template DNA.  
 57
  Because DNA recovery from PC extraction relies heavily on how careful a technician is to prevent phenol carry-through and involves multiple transfers, efficiency is variable. This explains why even though samples were pooled prior to extraction STR quality varied between samples. Because forensic exhibits are hard to come by, a more consistent protocol and DNA extraction method would be better than one that is highly variable. Further, it might be favourable to find an alternative method that avoids the use of harsh chemicals. A SCODA protocol would be advantageous since minimal sample handling is involved and organic extraction can be bypassed completely, as in the EDTASCODA protocol. An advantage to the SCODA protocol is that lysis can be performed directly on BSC, bypassing significant losses associated with swabbing. Since the extracts from this method are not sensitive to the high concentrations of calcium during PCR as is PC, overestimating the amount of stained concrete during collection could be acceptable. The exact maximum amount of concrete to stain ratio would need to be determined but this could be useful when a stain is difficult to visualize; investigators would want to sample a larger area to prevent evidence from being left behind. In this study, the author was able to show that chipping can be used as an alternative collection method to swabbing, especially in cases where low amounts of DNA are expected. Additionally, a sample preparation method to be used in conjunction with SCODA was developed. The centrifugal filtration step after lysis sufficiently reduces the sample conductivity to a level acceptable for SCODA. The chipping, followed by EDTA-SCODA approach also bypasses the need for samples to be processed by another extraction method, which could result in increased DNA losses.  
 58
  Chapter 5 – Casework 
 
  At the time that this project was conducted, two forensic cases were separately being examined in the BOLD Forensic Laboratory. In both cases, the goal was to obtain a DNA profile from collected material for identification purposes. Customary PC extraction failed for both samples and it was evident that the hindrance was caused by PCR inhibition. Having exhausted all traditional options, including re-purification and dilution, SCODA was applied as a final effort to see if the inhibition could be removed to yield an STR profile. These cases will be discussed in further detail here and in the following section.  5.1 A Murder Weapon 
 5.1.1 Introduction Investigators suspected that a particular object was used as a weapon in a murder. The object was seized from the crime scene and was analyzed for DNA by another laboratory prior to being submitted to the BOLD Laboratory for further assessment. Two profiles were anticipated to emerge, those from the victim and potentially from sloughed epithelial cells from the perpetrator who held the object. When the first laboratory examined the exhibit, analysts swabbed the bloodstained surfaced in an attempt to confirm that the DNA belonged to the victim. They then tested the other end that was potentially held by the perpetrator by washing its surface with moist gauze but it was the victim’s DNA that they recovered, presumably from small droplets of blood splatter that contaminated the weapon. The BOLD Laboratory then tried to see if it could improve results by soaking the object directly in lysis buffer to obtain as many epithelial cells as possible followed by PC extraction. Initially, PCR results indicated there was an inhibitor present in the sample, which failed  
 59
  to amplify. Upon two attempts at removing PCR inhibitors by a performing a second PC extraction and dilution, SCODA was applied and a mixed, partial DNA profile was acquired.  5.1.2 Methods and Materials Sample Collection and Lysis To ensure maximum DNA recovery and avoid losses through swabbing, based on results from retrieving trace DNA from bloodstains on concrete, the tip of the suspected weapon where touch-DNA was believed to have been was submersed in lysis buffer (10mM Tris-HCl pH=8.0, 10mM EDTA, 50mM NaCl, 2% SDS and Proteinase K, 0.4mg/ml). To avoid cross-contamination from the blood to touch-DNA as much as possible, a volume was chosen to cover the end opposite to the victim’s blood (Q1-1, Figure 1). However, as a precaution, a second volume of lysis buffer was chosen to cover a larger surface area in case more touch DNA could be retrieved; the second extraction (Q1-2) was performed after the lysate for Q1-1 had been collected. BIS11 and BIS12 (20 µl) and blanks were also prepared and extracted in the same manner as positive and negative controls, respectively.  Figure 5.1 – Illustration of the object processed twice with two different volumes.  Following an overnight incubation at 56oC for each extraction, the object was removed and the collected Q1-1 and Q1-2 lysates were diluted by a factor of 10 with  
 60
  FAD H2O. The diluted supernatants were then concentrated to 500µl in an Amicon Ultra15, 30KDa centrifugal filter (catalogue no. UFC903096) at 4 000g. The supernatants were each transferred to a 2ml heavy phase-lock gel and the procedure was continued as described in Chapter 4.2. SCODA and Speed-Vac Purification The samples eluted from PC-Round 2 were each diluted to 5ml with FAD H2O in the sample chamber of a disposable cartridge, the cartridges were loaded onto the Aurora, and the Aurora Forensic Sample Clean-up Protocol was run. In each case a 60µl output was collected and further concentrated by speed-vac concentration (Eppendorf, Vacufuge model 5301) at room temperature. Finally, the lyophilized samples were re-suspended in 15µl TE buffer for quantification and STR amplification. DNA Analysis Quantification, STR amplification and analysis of DNA were all carried out as previously described in Chapter 3.2.  5.1.3 Results In Q1-1 and Q1-2, quantification of DNA was either below the standard curve or non-existent in undiluted PC extracts and those diluted by 1:5. Amplification was however possible after being purified with SCODA (Figure 5.2a). Q1-1 PC-Round 1 showed signs of inhibition where the undiluted 1:1 output did not amplify and the 1:5 dilution amplified more than 5 cycles after the standards of that run (Figure 5.2b). Q1-1 PC-Round 2 is no longer inhibited at 1:5, but remains inhibited at 1:1. The same extract was then purified with SCODA and subjected to speed-vac for concentration (SCODA-spVac), which made amplification possible at 1:1. A 1:5 dilution after SCODA was not performed due to the concern that limited amounts of material would be available for STR amplification. Similar results were seen in Q1-2, where the  
 61
  sample extracted by PC is inhibited at 1:1 but not after the sample is cleaned with SCODA-spVac. These results are consistent with the quantification results.  Total DNA (ng)  a. DNA Recovery - Quantifiler 0.4 0.3 0.2 0.1 0  1:1 PCRound1  PCRound2  SCODA  Q1-1  PCRound1  SCODA  1:5  Q1-2  
  IPC Delay (# of cycles)  b. PCR Inhibition - IPC 30 25 20 15 10  1:1  5  1:5  0 PC-Round1 PC-Round2 Q1-1  SCODA  PC-Round1  SCODA  Q1-2  Figure 5.2 – Quantifiler™ Human DNA Quantification Results. (a) Quantifiler™ human DNA quantities of Q1-1 from two PC extracts (PC-Round 1 and 2) and SCODA followed by speed-vac (SCODA). Undiluted samples are denoted as 1:1, while 1:5 indicates that the sample was diluted by a factor of 5. (b) PCR inhibition represented as the number of IPC CT delays from averaged IPC CT values of DNA standards.  
 62
  Figure 5.3 is the mixed, partial profile that was obtained from sample Q1-1. To ensure that this was not result from cross contamination during extraction or amplification and that SCODA was successful, profiles of BIS11, BIS12, PCR positive, and negative controls were assessed. Additionally, it was also compared to DNA profiles from the internal laboratory database of present and past employees, visitors and past cases. Extraction positives: BIS11 and BIS12, and PCR positive samples gave expected profiles, while the negative controls gave no profiles and cross-contamination from exogenous DNA was also ruled out.  
 Figure 5.3 – STR Profile Obtained from SCODA-spVac To protect the identity of those involved in the investigation the scale and range in RFUs and allele calls were removed. 
 Once it was determined that the profile was not a result of cross-contamination, it was compared to profiles belonging to the victim and several persons of interest that were supplied by investigators. Although the minor contributor could not be conclusively identified, the major contributor was determined to be the victim. Amplification was also  
 63
  attempted for 1:5 and 1:10 dilutions of the Q1-1 PC extracts but no profile peaks were apparent.  5.1.4 Discussion The main aim of this study was to see if enough epithelial cells could be extracted from the object to identify the perpetrator in this forensic case. Although the major contributor from the obtained partial profile belonged to the victim, the idea that SCODA can be used for purifying inhibited forensic samples was realized in this study. The object in this section was the perfect candidate for SCODA, exhibiting a high concentration of inhibitors that even when diluted, mask the low amounts of DNA. Organic extraction was used to extract DNA from a large volume of lysate and when two attempts at obtaining a profile using this method failed, SCODA was able to clean the extract—it was not the profile investigators were looking for but it was a profile nonetheless. One recurring outcome that was seen in this study with Q1-2 and in the study with BB9 (Chapter 3) was that a profile could not be obtained even though quantification indicated sufficient amounts of DNA were present. In Chapter 3, it was mentioned that this discrepancy could be due to the higher sensitivity of the STR amplification process to inhibitors relative to the QuantifilerTM reaction due to the larger target sizes and number of primer pairs used in STR amplification assays. An examination of the object itself, how it was stored and the fact that it had been processed twice before also makes it probable that there was not a sufficient amount of high quality DNA, that it was too degraded or fragmented for amplification by STR primers. To ensure that the opposite case was not true, where quantification failed but an STR profile could be obtained, the 1:5 and 1:10 dilution of Q1-1 as extracted by PC was also subjected to amplification but no peaks were apparent. These extracts were likely too inhibited or contained an inadequate amount of target DNA for STR amplification.  
 64
  5.2 Human Remains 
 5.2.1 Introduction Skeletal remains were found in a drained water reservoir, which normally would have been covered by up to six metres of water. At the time of the investigation, there had only been one report of a missing person within the area; therefore, a positive identification was required to see if the human remains belonged to the missing person. The goal was to see if a usable DNA profile could be obtained from the remains, first through the use of standard BOLD Laboratory procedures and then with SCODA. Due to the sensitive nature of this investigation, only the profile obtained from the remains will be shown.  5.2.2 Methods and Materials 
 To maximize the amount of extracted DNA, three sections of the selected hard tissue exhibit were subjected to cryogenic grinding and PC extraction, which was carried out as described in Chapter 3.2. Extractions were conducted as separate reactions until the concentration step in a Vivacon-500 ETO where they were combined; the sample in question will be referred to as Q. Extraction positive (BIS10) and negative controls were included. After the first PC extract had failed to quantify, the sample was divided into two aliquots. One aliquot was concentrated and washed again in a new Vivacon-500 ETO using two 400µl volumes, while the other aliquot was subjected to SCODA purification and speed-vac concentration as outlined in Chapter 5.2. Quantification, and STR amplification and analysis were conducted as first described in Chapter 3.2. 
  
 65
  5.2.3 Results Figure 5.4 shows that Q is completely inhibited and appears as a false negative after PC extraction with only one wash (PC-Wash 1). After carrying out a second wash (PC-Wash 2), the undiluted PC extract still remains inhibited and although a dilution at 1:10 clears the inhibition, the sample quantifies below the standard curve but STR amplification was attempted anyway. In the aliquot that was purified with SCODA followed by speed-vac (SCODA), approximately 0.15ng/µl was recovered however; IPC indicates that it is slightly inhibited. As a precaution, the SCODA extract was diluted by a factor of 5 and 10 for STR amplification in addition to performing PCR on the undiluted extract. Figures 5.5 and 5.6 show the partial STR profiles that were obtained from PCWash 2 diluted by 1:10 and SCODA diluted by 1:5, respectively; note the difference in the RFU scale. From the other SCODA samples that were amplified, the undiluted failed to produce any peaks, while the output diluted by 1:10 had less allele calls than shown in Figure 5.6.  
 66
  
  
  b. Inhibition - IPC IPC Delay (# of cycles)  30 25 20 15  1:1  10  1:10  5 0 PC-Wash 1  BIS10  Ext.Neg.  PC-Wash 2  SCODA  
  ™  Figure 5.4 – Quantifiler Human DNA Quantification Results. (a) DNA recovery of Q from PC extraction (followed by 1 and 2 washes with a Vivacon500) and SCODA with speed-vac (SCODA). (b) PCR inhibition represented as the number of IPC CT delays from averaged IPC CT values from DNA standards.
 
  
 67
  
 Figure 5.5 – PC extracted STR Profile of Q. 
  
 68
  
 Figure 5.6 – SCODA extracted STR Profile of Q. Circled in red are peaks that are below threshold (50RFU) that may be enhanced and likely confirmed to be actual alleles if the injection time during CE was increased. 
 5.2.4 Discussion  As with the murder weapon, the skeletal exhibit presented in this section proved to be a prime candidate for the application of SCODA. If SCODA purification had not been attempted, a profile yielding six of nine usable loci may have not been obtained. One environmental factor that could have contributed to this success was if the remains were once submerged in water from the reservoir. It has been shown that water can preserve DNA from degradation 45. If the DNA had been heavily degraded, a profile may not have been obtained even if SCODA could remove PCR inhibition.  
 69
  An interesting result that emerged was how the undiluted SCODA output was inhibited and a profile could only be obtained after the extract was diluted by a factor of five. This indicates that although a majority of the inhibition had been removed, a residual amount still affected PCR to the extent that it was completely inhibited. Further, speed-vac was applied post SCODA to gain a three-fold increase in DNA concentration; if this step was not performed, the concentration of inhibitors would have been less by the same magnitude. A significant result presented in this section was that SCODA removed a majority of inhibitors from a sample that otherwise would not yield a DNA profile. It might also be possible to obtain a full profile if the injection time during CE was increased, which was beyond the scope of this particular case. A limitation with SCODA that was realized in this study also is that further improvements on inhibitor removal to completely purify the DNA without having to further dilute the extract is needed.  
 70
  Chapter 6 – Conclusions In this project, two possible applications of SCODA in forensic science have been presented. In one demonstration, the author has shown that SCODA can be used as a secondary purification step to remove inhibitors that have carried through by using another extraction method such as phenol-chloroform. Particularly, this was successfully executed on the murder weapon and the skeletal remains that came into the BOLD Laboratory in Chapter 5, where SCODA was able to recover some partial profiles after traditional approaches failed. These results provide the evidence to prove that samples with an extremely large concentration of inhibitors to relative DNA yield do exist and in these cases SCODA would be extremely valuable. Having shown that SCODA can successfully be used as a secondary purification process to other extraction methods, it was the author’s inclination to develop a preSCODA sample prep that would bypass other extraction methods like PC, to prevent losses and time associated with multiple handling steps. A direct-to-SCODA sample lysis procedure has always been a challenge due to the sample conductivity limit that SCODA has, where injection is efficient only if the sample conductivity is 25% of the gel conductivity. However, through the use of a quick centrifugal filtration step, the sample can be desalted so that the conductivity requirement can be achieved. Therefore, the second demonstration of SCODA was to show that a (semi) direct to SCODA protocol could be developed, in this case, to address the problems associated with obtaining DNA from bloodstains on concrete. Although comparable yields were obtained from using EDTA lysis coupled with SCODA and PC extraction, further optimization with the protocol could better improve SCODA yields and conducting more experimental replicates may indicate higher yields for one method over the other. Aside from the SCODA successes, it was also shown that chipping can be used as an alternative to swabbing stains on concrete to obtain more DNA.  
 71
  An outcome that has been encountered throughout this project is the difficulty in amplifying the largest loci: FGA, D18S51, and D7S820. This is not surprising to see since it is common of heavily degraded samples. For successful PCR amplification, the DNA template must be intact. If there is a break in the template, primer extension will stop when it reaches the break. Many studies have shown that the larger the target region, the more vulnerable it is to being fragmented 6. Although STR analysis is a dramatic improvement from RFLP analysis in terms of targeting smaller regions of DNA to overcome degradation, it is inevitable that some extremely degraded samples will still fail. In an effort to increase the success in obtaining profiles from degraded samples, companies like Applied Biosystems have redesigned their primer sets to target flanking DNA that are more closely situated near the STR regions to reduce the size of amplicons. This newer technology is known as miniSTRs. On a larger scale, the fact that specific SCODA protocols can be designed to, not only meet SCODA requirements but also tackle samples based on particular problems they present to downstream processes, only allows for SCODA to have more of an advantage over other extraction methods, which are more general and do not necessarily target the specific sample-type problems. Future work with SCODA in forensics will include the development of more protocols for samples such as bone, teeth, and stains on denim, touch-DNA and also the improvement of SCODA’s sample conductivity tolerance and progression away from the necessity to desalt. Finally, other exciting endeavours in forensic science will even include work with sequence specific SCODA, a new generation of SCODA platforms that have the ability to enrich for target DNA sequences with distinguishing capabilities down to 1 nucleotide. Imagine being able to select for all the STR sequences from a mixture of multiple DNA sources in one sample; this could potentially be a very powerful tool for processing samples like stains in soil or post-blast bomb fragments where bacterial background DNA is abundant or trace amounts of highly degraded DNA are present. For difficult, inhibited samples that otherwise do not yield a DNA profile, SCODA could potentially be the solution to these and be an asset to the forensic community.  
 72
  References 1. Rudin, N. & Inman, K. An Introduction to Forensic DNA Analysis, Second Edition. (CRC Press: 2001). 2. Maze, M., Stagnara, D. & Fischer, L.-P. [Dr. Edmond Locard (1877-1966), the Sherlock Holmes of Lyons]. Hist Sci Med 41, 269-278 (2007). 3. Butler, J.M. Forensic DNA typing. (Academic Press: 2005). 4. Alberts, B. Molecular Biology of the Cell 5th Edition HARDCOVER. (Garland Science: 2007). 5. Butler, J.M. Genetics and Genomics of Core Short Tandem Repeat Loci Used in Human Identity Testing. J Forensic Sci 51, 253-265 (2006). 6. National DNA Data Bank. The First Canadian Conviction with DNA Evidence. (2001). <http://www.nddb-bndg.org/cases/mcnally_e.htm> 7. Butler, J.M. Fundamentals of Forensic DNA Typing. (Academic Press: 2009). 8. Chayko, G.M. & Gulliver, E.D. Forensic Evidence in Canada. (Canada Law Book Inc.: Ontario, Canada, 1999). 9. Jobling, M.A. & Gill, P. Encoded evidence: DNA in forensic analysis. Nat Rev Genet 5, 739-751 (2004). 10. Thompson, J., Kibler, J. & Kupferschmid, T. Comparing Five Forensic DNA Extraction Methods. (2007). 11. Sambrook, J., MacCallum, P., & Russell, D. (Professeur Molecular cloning: a laboratory manual. (CSHL Press: 2001). 12. Valgren, C., Wester, S. & Hansson, O. A comparison of three automated DNA purification methods in Forensic casework. Forensic Science International: Genetics Supplement Series 1, 76-77 (2008). 13. Rohland, N. & Hofreiter, M. Ancient DNA extraction from bones and teeth. Nat. Protocols 2, 1756-1762 (2007). 14. Marziali, A., Pel, J., Bizzotto, D. & Whitehead, L.A. Novel electrophoresis mechanism based on synchronous alternating drag perturbation. ELECTROPHORESIS 26, 82-90 (2005).  
 73
  15. Green, R.L., Roinestad, I.C., Boland, C. & Hennessy, L.K. Developmental validation of the quantifiler real-time PCR kits for the quantification of human nuclear DNA samples. J. Forensic Sci 50, 809-825 (2005). 16. Johns, L.M., Thakor, A., Ioannou, P., Kerai, J. & Thomson, J.A. Validation of Quantifiler(TM) Human Quantification Kit for forensic casework. International Congress Series 1288, 762-764 (2006). 17. Koukoulas, I., O’Toole, F.E., Stringer, P. & van Oorschot, R.A.H. Quantifiler observations of relevance to forensic casework. J. Forensic Sci 53, 135-141 (2008). 18. Roby, R. & Figarelli, D. Capillary Electrophoresis. (2009). <http://www.nfstc.org/pdi/Subject05/pdi_s05_m02.htm> 19. Figarelli, D. Data Interpretation and Allele Calls. (2009). <http://www.nfstc.org/pdi/Subject06/pdi_s06_m03.htm> 20. Butler, J.M. SWGDAM Autosomal STR Interpretation Guidelines. (2010). 21. SWGDAM Interpretation Guidelines for Autosomal STR Typing by Forensic DNA Testing Laboratories. (2010). 22. Bessetti, J. An Introduction to PCR Inhibitors. (2007). 23. Eilert, K.D. & Foran, D.R. Polymerase Resistance to Polymerase Chain Reaction Inhibitors in Bone. Journal of Forensic Sciences 54, 1001-1007 (2009). 24. Broemeling, D., Pel, J., Gunn, D., Mai, L., Thompson, J., Poon, H. & Marziali, A. An Instrument for Automated Purification of Nucleic Acids from Contaminated Forensic Samples. Journal of the Association for Laboratory Automation 13, 40-48 (2008). 25. Pel, J., Broemeling, D., Mai, L., Poon, H., Tropini, G., Holt, R.A. & Marziali, A. Nonlinear electrophoretic response yields a unique parameter for separation of biomolecules. Proceedings of the National Academy of Sciences 106, 14796-14801 (2009). 26. Boreal Genomics Aurora User Manual. (2010). 27. Loreille, O.M., Diegoli, T.M., Irwin, J.A., Coble, M.D. & Parsons, T.J. High efficiency DNA extraction from bone by total demineralization. Forensic Sci Int Genet 1, 191-195 (2007). 28. Lee, H.Y., Park, M.J., Kim, N.Y., Sim, J.E., Yang, W.I. & Shin, K. Simple and highly effective DNA extraction methods from old skeletal remains using silica columns. Forensic Science International: Genetics 4, 275-280 (2010).  
 74
  29. Sutlovic, D., Gamulin, S., Definis-Gojanovic, M., Gugic, D. & Andjelinovic, S. Interaction of humic acids with human DNA: Proposed mechanisms and kinetics. Electrophoresis 29, 1467-1472 (2008). 30. Opel, K.L., Chung, D. & McCord, B.R. A Study of PCR Inhibition Mechanisms Using Real Time PCR. Journal of Forensic Sciences 55, 25-33 (2010). 31. Robe, P., Nalin, R., Capellano, C., Vogel, T.M. & Simonet, P. Extraction of DNA from soil. European Journal of Soil Biology 39, 183-190 (2003). 32. Poon, H. Personal Communication. (2010). 33. Handt, O., Höss, M., Krings, M. & Pääbo, S. Ancient DNA: Methodological challenges. Experientia 50, 524-529 (1994). 34. Soulsbury, C.D., Iossa, G., Edwards, K.J., Baker, P.J. & Harris, S. Allelic dropout from a high-quality DNA source. Conserv Genet 8, 733-738 (2006). 35. Barbaro, A., Cormaci, P. & Barbaro, A. Validation of DNA typing from skeletal remains using the Invitrogen Charge Switch® Forensic DNA Purification Kit. Forensic Science International: Genetics Supplement Series 1, 398-400 (2008). 36. Horsman-Hall, K.M., Orihuela, Y., Davis, A.L., Ban, J.D. & Greenspoon, S.A. Development of STR profiles from firearms and fired cartridge cases. Forensic Science International: Genetics 3, 242-250 (2009). 37. van Oorschot, R.A., Ballantyne, K.N. & Mitchell, R.J. Forensic trace DNA: a review. Invest Genet 1, 14 (2010). 38. Westen, A.A., Nagel, J.H., Benschop, C.C., Weiler, N.E., de Jong, B.J. & Sijen, T. Higher Capillary Electrophoresis Injection Settings as an Efficient Approach to Increase the Sensitivity of STR Typing. Journal of Forensic Sciences 54, 591-598 (2009). 39. Linke, D. Chapter 34 Detergents: An Overview. Guide to Protein Purification, 2nd Edition Volume 463, 603-617 (2009). 40. Sigma-Aldrich Detergent Properties and Applications. BioFiles 3, 14 (2008). 41. Swango, K.L., Timken, M.D., Chong, M.D. & Buoncristiani, M.R. A quantitative PCR assay for the assessment of DNA degradation in forensic samples. Forensic Sci. Int 158, 14-26 (2006). 42. Hewlett, P. Lea’s Chemistry of Cement and Concrete, Fourth Edition. (ButterworthHeinemann: 2004).  
 75
  43. van Oorschot, R.A.H., Phelan, D.G., Furlong, S., Scarfo, G.M., Holding, N.L. & Cummins, M.J. Are you collecting all the available DNA from touched objects? International Congress Series 1239, 803-807 (2003). 44. Sweet, D., Lorente, M., Lorente, J.A., Valenzuela, A. & Villanueva, E. An improved method to recover saliva from human skin: the double swab technique. J. Forensic Sci 42, 320-322 (1997). 45. Lowe, C.M. Forensic DNA Identification from Human Remains Submerged in Water. (2009). <http://www.slideshare.net/livestrong8421/forensic-dnaidentification-from-human-remains-submerged-in> 46. Welch, B.L. The Generalization of "Student's" Problem When Several Different Population Variances are Involved. Biometrika 34, 28 -35 (1947).  
 76
  Appendix A – Methods and Materials Sample Size In forensic science, especially where trace evidence is being analzed, examiners often do not have the luxury of splitting the sample into replicates. To be confident that an obtained DNA profile is the true profile belonging to a given bone, a swabbed item, or a victim, etc… a host of laboratory controls are conducted and analyzed in the same manner as the DNA in question. Therefore, because the bones and casework exhibits processed in Chapters 3 and 5, respectively, were precious samples, replicates were not included. However, high confidence can be had since the necessary controls were all valid. These controls include an extraction positive and negative, and an amplification positive and negative. In Chapter 4, where mocked BSC samples were analyzed, the Welch’s t test, a statistical calculation, was performed to determine whether EDTA-SCODA was significantly better than the next best method, PC extraction. This test is used when the true mean and variance of a population is unknown and two types of samples have unequal variances, s 46. From the data set obtained, EDTA-SCODA extraction on BSC had sample mean, Χ =0.371 and s=0.250, while for PC extraction, Χ =0.123 and s=0.078. Welch's t-test defines the statistic t by the following formula: € €  where Χ i, s2i and Ni are the ith sample mean, sample variance, and sample size respectively. The degrees of freedom, ν, associated with this variance estimate is €  approximated using:  
 77
  here, νi = Ni − 1, the degrees of freedom associated with the ith variance estimate. In both cases N=9, which means that the ν=8. Using the formulas, and reported sample mean and variance of the extraction methods above, the calculated t=2.833 and ν=9.520 were obtained. And according to the t-table following a one-tail distribution, EDTA-SCODA is shown be significantly better than PC extraction, p=0.01.  STR Analysis Where STR data is presented on the experimental samples, it can be assumed that controls (allelic ladder, PCR positive, and PCR negative) were analyzed first and appeared as expected, prior to analyzing samples. SCODA Gel and Running Buffer On the HsBB, all reusable SCODA gelboats were soaked in 10% bleach for 1hr, rinsed with nanopure H2O and irradiated with UV light prior to casting the gel. In all experiments, including those found in disposable cartridges, 1% Seakem LE agarose gels (Lonza) made with 0.25x Tris Borate EDTA buffer pH=8.0 (1xTBE, Ambion) were used; the same strength TBE was used as running buffer. Its conductivity is ~1200 µS/cm. SCODA Run Conditions All experiments carried out on the Aurora were run with the Aurora DNA Cleanup protocol; details are listed in Table A.1.  
 78
  
  Platform  HsBB and Aurora Instruments  Agarose Gel  4.5 mm 1% Seakem LE  Running Buffer  0.25x TBE  Elution Volume  60 µl  Injection  600 V @ 3 000 mC  Wash  70 V/cm, 4 s period, 20% Bias, 2.5 hours  Focus  70 V/cm, 4 s, period, 1.5 hours  Total Run Time  4 hours  Table A.1 – SCODA Run Conditions.  
 79
  

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.24.1-0105145/manifest

Comment

Related Items