UBC Faculty Research and Publications

Novel genomic resources for a climate change sensitive mammal: characterization of the American pika… Lemay, Matthew A; Henry, Philippe; Lamb, Clayton T; Robson, Kelsey M; Russello, Michael A May 10, 2013

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

Item Metadata


52383-12864_2012_Article_5020.pdf [ 451.41kB ]
JSON: 52383-1.0223378.json
JSON-LD: 52383-1.0223378-ld.json
RDF/XML (Pretty): 52383-1.0223378-rdf.xml
RDF/JSON: 52383-1.0223378-rdf.json
Turtle: 52383-1.0223378-turtle.txt
N-Triples: 52383-1.0223378-rdf-ntriples.txt
Original Record: 52383-1.0223378-source.json
Full Text

Full Text

Novel genomic resources for a climate changesensitive mammal: characterization of theAmerican pika transcriptomeLemay et al.Lemay et al. BMC Genomics 2013, 14:311http://www.biomedcentral.com/1471-2164/14/311RESEARCH ARTICLE Open AccessNovel genomic resources for a climate changesensitive mammal: characterization of theAmerican pika transcriptomeMatthew A Lemay1*, Philippe Henry1,2, Clayton T Lamb1, Kelsey M Robson1 and Michael A Russello1AbstractBackground: When faced with climate change, species must either shift their home range or adapt in situ in orderto maintain optimal physiological balance with their environment. The American pika (Ochotona princeps) is a smallalpine mammal with limited dispersal capacity and low tolerance for thermal stress. As a result, pikas have becomean important system for examining biotic responses to changing climatic conditions. Previous research usingamplified fragment length polymorphisms (AFLPs) has revealed evidence for environmental-mediated selection inO. princeps populations distributed along elevation gradients, yet the anonymity of AFLP loci and lack of availablegenomic resources precluded the identification of associated gene regions. Here, we harnessed next-generationsequencing technology in order to characterize the American pika transcriptome and identify a large suite ofsingle nucleotide polymorphisms (SNPs), which can be used to elucidate elevation- and site-specific patterns ofsequence variation.Results: We constructed pooled cDNA libraries of O. princeps from high (1400m) and low (300m) elevation sitesalong a previously established transect in British Columbia. Transcriptome sequencing using the Roche 454 GS FLXtitanium platform generated 780 million base pairs of data, which were assembled into 7,325 high coveragecontigs. These contigs were used to identify 24,261 novel SNP loci. Using high resolution melt analysis, wedeveloped 17 of these SNPs into genotyping assays, which were validated with independent DNA samples fromBritish Columbia Canada and Oregon State USA. In addition, we detected haplotypes in the NADH dehydrogenasesubunit 5 of the mitochondrial genome that were fixed and different among elevations, suggesting that this maybe an informative target gene for studying the role of cellular respiration in local adaptation. We also identifiedcontigs that were unique to each elevation, including a high elevation-specific contig that was a positive matchwith the hemoglobin alpha chain from the plateau pika, a species restricted to high elevation steppes in Asia.Elevation-specific contigs may represent candidate regions subject to differential levels of gene expression alongthis elevation gradient.Conclusions: To our knowledge, this is the first broad-scale, transcriptome-level study conducted within theOchotonidae, providing novel genomic resources for studying pika ecology, behaviour and population history.Keywords: Adaptation, Elevation gradient, Next-generation sequencing, Ochotona princeps, Population genomics,Single nucleotide polymorphisms* Correspondence: matt.lemay@ubc.ca1Department of Biology, University of British Columbia, Okanagan Campus3333 University Way, Kelowna, BC V1V 1V7, CanadaFull list of author information is available at the end of the article© 2013 Lemay et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.Lemay et al. BMC Genomics 2013, 14:311http://www.biomedcentral.com/1471-2164/14/311BackgroundWhen faced with rapidly changing climates, many speciesare expected to undergo widespread shifts in theirdistribution in order to maintain optimal physiologicalbalance with their environment [1]. However, for specieswith fragmented habitats and those with limited dispersalcapacities, range shifts may not be a viable option, andrapid adaptation may represent the only alternative to localextinction [2,3]. In order to predict the ability of thesespecies to evolve in situ to changing environmentalconditions, studies examining local adaptation alongelevation gradients have emerged as model systems topredict the impact of climate change on species persistenceand survival [4].The American pika (Ochotona princeps) is a small alpinelagomorph with a discontinuous distribution throughoutthe mountain ranges of western North America [3,5].Pikas are typically restricted to high-elevation talus slopeecosystems, which provide close proximity to meadowsfor foraging and a complex habitat for behaviouralthermoregulation [6,7]. American pikas likely originatedfrom an Asian ancestor that arrived in North Americavia the Bering land bridge [8]. During the warming thatfollowed the Wisonsinan glaciation, paleontological evi-dence suggests that the distribution of O. princepscontracted northward and to higher elevations [9],effectively stranding extant populations on high-elevation‘habitat islands’. Currently, the lower limits of O. princepspopulations are constrained by an inability to toleratethermal stress, while their high elevation distribution isenabled by adaptation to hypoxic environments [10]. Theuniquely fragmented nature of their habitat has propelledO. princeps to a focal mammalian species for more generalstudies of metapopulation dynamics, island biogeography,and source-sink dynamics [9,11].Pikas have also emerged as an important study speciesfor investigating extinction risk in the face of rapidlychanging climates [5,7,12-15]. Unlike the majority ofwoodland montane fauna whose continuous habitatallows for cross-valley dispersal among mountain ranges,pikas reliance on high-elevation talus habitat precludestheir ability for dispersal to cooler latitudes [9]. Instead,it is hypothesized that the continued persistence of pikaswill depend on in situ adaptation to changing climaticconditions, leading some to suggest that they maybecome the first mammalian species to go extinct due tothe direct effects of climate change [16]. Investigatingthe genetic basis of adaptation in pikas may provideinsight into the underlying mechanisms by whichcontemporary evolution occurs in response to rapidlychanging environments. However, this research is hinderedby a lack of available genomic resources. For example, arecent genome scan using amplified fragment lengthpolymorphisms (AFLPs) among populations continuouslydistributed along three elevation gradients (0 m-1500 m)identified 15 outlier loci (out of 1509) putatively exhibitingsignatures of divergent selection associated with summermean maximum temperature and precipitation (PhilippeHenry and Michael Russello, unpublished data). Yet, theanonymity of AFLP loci precluded the identification ofunderlying genomic regions associated with these candi-date loci.The rise of next-generation sequencing technologiesprovides tools for rapidly generating DNA sequencedata for non-model organisms that have previouslylacked genomic resources. When combined with statisticalpopulation genomics approaches [17], these data can beused to test for signatures of natural selection inwild populations and identify candidate gene regionsassociated with local adaptation [18]. Single nucleotidepolymorphisms (SNPs) have emerged as the marker ofchoice for population-level genotyping in the genomicsera [19,20]. Due to their high coverage across the genome,ease of genotyping, and direct relationship with underlyinggene function, SNPs represent an improvement overconventional markers such as AFLPs and microsatellitesfor identifying genome-wide patterns of adaptive geneticvariation [21]. Despite their utility for population levelstudies, large-scale SNP resources are still lacking formany species, including O. princeps.The purpose of this study was to harness next-generationsequencing technology in order to elucidate elevation-specific patterns of sequence variation in O. princeps. Wegenerated transcriptome-wide sequence data for pooledcDNA libraries from high (1400 m) and low (300 m)elevation sites along a previously established elevationgradient in the British Columbia (BC) Coast Mountains[13]. The resulting high coverage contigs and large suite ofSNP loci represent novel genomic resources for studyingpika ecology, behaviour and population history, and enabledirect investigations of potential biotic responses tochanging environments.Results and DiscussionSequencing and assemblyUsing the Roche 454 GS FLX titanium platform, wegenerated ~780 million bases of transcriptome sequencedata corresponding to 1.6 × 106 and 1.5 × 106 reads forthe high and low elevation cDNA libraries, respectively(Table 1).Table 1 Summary of the next-generation sequence dataobtained from each cDNA libraryHigh elevation Low elevationNo. of bases 424,134,294 357,655,427No. of reads 1,589,727 1,455,497Mean read length 266.8 245.7Lemay et al. BMC Genomics 2013, 14:311 Page 2 of 11http://www.biomedcentral.com/1471-2164/14/311A de novo assembly was first carried out using thetrimmed reads from both elevations in order to generatereference contigs; this assembly incorporated 66% of thetranscriptome reads to produce 102,175 contigs. Wethen mapped the raw reads back to these referencecontigs separately for each elevation in order to generatea refined dataset consisting only of contigs that had aminimum average coverage of 5× for each elevation anda minimum length of 200 bases. The resulting dataset(hereafter referred to as the high coverage dataset)consisted of 7,325 contigs with a mean coverage of 33reads per site (Table 2; Figure 1; Additional files 1 and 2).These contigs represent less than 1% of the O. princepsgenome, which initial low coverage estimates indicate is1.92 Gb in length [22].We performed an additional de novo assembly (simi-larity = 0.90) of the high coverage contigs in order toidentify sequences that either partially or totally overlapped.This assembly revealed some redundancy in the contigdataset. Out of the 7,325 contigs in the high coveragedataset, 588 contigs (8.0%) aligned with one other contig,and 221 contigs (3.0%) aligned with two or more othercontigs. The remaining 6,516 contigs (89.0%) were uniqueand did not show similarity with any other contig.Transcriptome annotationA BLAST search of all contigs in the high coverage dataset(7,325 sequences) produced 3,788 positive hits (BLASTxsearch of the NCBI nr database, minimum e-value cutoff = 10-6; average e-value = 3.4 × 10-9; Additional file 2).Of the positive BLAST hits, only 14 were matches tosequences from Ochotona sp., highlighting the currentlack of genomic resources available for pikas; 1,215contigs had positive matches to published genes fromthe European rabbit (Oryctolagus cuniculus), which isthe closest model organism to O. princeps. Of the contigsequences with positive BLAST match, 2,279 weresubsequently annotated with one or more gene ontology(GO) terms (Figure 2).SNP detectionAmong the high coverage contigs (n = 7,325), 5,357 hadSNPs that fell within our detection parameters (Additionalfile 3). The total number of SNPs identified was 24,261, ofwhich 3,399 were polymorphic among pika from bothelevations, 10,504 were polymorphic in low elevation butfixed in high elevation pika, and 10,269 were polymorphicin the high elevation but fixed in low elevation. There were89 SNPs within our detection parameters that appeared tobe fixed for alternate alleles in the two elevations. The ratioof transitions to transversions was 3.86, and the differencein the frequency of the major allele between the two eleva-tions ranged from <1 to 100% (mean divergence = 21%).Among these data, the frequency of SNPs that appearto be fixed at one elevation may be artificially inflateddue to the small sample size (n = 3 for each elevation)used to generate the transcriptome sequences. There is ahigh probability that low frequency alleles would nothave been present among the individuals sampled. Inaddition, the SNP detection parameters required aminimum coverage of eight reads at a polymorphic site tobe included in the data. If the samples from one elevationhad low coverage at a particular site, it would appear to befixed even if there was variation present. These potentialbiases reflect the trade-off between avoiding false SNPsresulting from sequencing error, while attempting toaccount for all possible variation in the data.SNP validationPrimer pairs were designed for 85 SNP loci such that theyamplified an ~200 base pair fragment that containeda single SNP (Additional file 4). Of these loci, 26 hadsuccessful PCR amplification, were free of introns, andproduced sufficiently clear high resolution melt (HRM)signal to attempt the subsequent genotyping validation.High resolution melt analysis was then used to genotype10 high and 11 low elevation O. princeps from the BellaCoola, BC study site as well as 21 samples collected atan independent location in the Columbia River Gorge,Oregon, USA.Sanger re-sequencing of representative samples fromeach melt curve obtained from these 26 loci was used toassign genotypes to each cluster. From the panel of 26SNPs for which Sanger validation was attempted, 17 loci(65%) yielded evidence of consistently scorable nucleotidepolymorphism. Sanger sequence data for the remainingnine loci confirmed that the expected SNP site was indeedpolymorphic, however, the resulting HRM curves wereTable 2 Summary of the contigs present in each O. princeps datasetTotal contigs High coverage dataset1 High elevation unique contigs2 Low elevation unique contigs2No. of contigs 102,175 7,325 1,038 304Mean coverage 5.5 33.2 7.4 8.0Mean length 534.6 1,079.5 383.2 354.1Mean no. of reads 16.6 137.0 11.4 12.91 In the high coverage dataset each contig has a minimum length of 200 bases and a minimum of 5× coverage for each ecotype.2 Contigs composed of reads from a single elevation. Minimum length = 200 bases; minimum coverage = 5×.Lemay et al. BMC Genomics 2013, 14:311 Page 3 of 11http://www.biomedcentral.com/1471-2164/14/311not sufficiently discrete to enable accurate genotypeassignment (i.e. Sanger sequencing revealed that multiplemelt curves had the same genotype or identified multiplegenotypes within the same cluster). Given that in thesecases Sanger sequencing confirmed the presence ofthe expected polymorphism, we conclude that the failedassays were not due to errors with the initial SNP detectionbut rather reflect the limitations of the HRM assaysat those loci. For example, the presence of additionalpolymorphic sites within the amplicon [23] and locicontaining Class 3 (C/G) or Class 4 (A/T) SNPs [24] mayresult in complicated or weakly differentiated clustersunsuitable for HRM genotyping.Eight of these 17 retained SNP loci exhibited sequencesimilarity to structural or regulatory genes in the NCBIdatabase (Ocp4162, Ocp6361, Ocp6774, Ocp7498,Ocp14764, Ocp15508, Ocp17339, Ocp102175; Additionalfile 2). We found no evidence of linkage disequilibriumamong any of the loci that were successfully typed in oursamples. Four of 17 loci showed a significant deviationfrom Hardy-Weinberg equilibrium (HWE), however eachinstance was restricted to a single elevation at one location(Table 3).All 17 loci tested were polymorphic among the 21DNA samples from BC. Four of these loci were fixed fora single allele at high elevation and five loci were fixedfor a single allele at low elevation (Table 3), potentiallyindicating elevation-specific patterns in the distributionof genetic variation. The remaining eight loci werepolymorphic at both elevations in BC.Among the DNA samples from Oregon, six loci weremonomorphic. Of the remaining 11 loci, four were fixedfor a single allele at the high elevation and two were fixedfor a single allele at the low elevation site. Reduced geneticvariation in samples from Oregon is likely representativeof ascertainment bias (and low sample sizes), given thattranscriptome sequencing and initial SNP discoveryutilized tissue samples from BC.Mitochondrial DNAThere was a high coverage of reads across all genes inthe O. princeps mitochondrial genome [GENBANK:AJ537415], with 11,040 trimmed reads (0.4%) aligning tothe published reference sequence. In addition, a BLASTxsearch of the high coverage dataset revealed 103 contigsthat associated with the mitochondria.Of particular note, we detected multiple SNPs within twocontigs (contigs 1829 and 24554) that sequence-similaritysearches revealed corresponded to portions of theNADH dehydrogenase subunit 5 (ND5) region of themitochondrial genome. Two distinct haplotypes weredetected across a total of eight polymorphic sites thatFigure 1 Characterization of contigs present in the high coverage dataset. Histograms represent (A) average coverage of each contig(mean = 33×), (B) number of reads that mapped to each contig (mean = 137.0), (C) contig lengths (mean = 1079.5 base pairs), and (D) thenumber of SNPs for each of the high coverage contigs.Lemay et al. BMC Genomics 2013, 14:311 Page 4 of 11http://www.biomedcentral.com/1471-2164/14/311associated with elevation in BC (Figure 3). Three of thesepolymorphic sites were non-synonymous substitutions,two of which occurred in loop regions, while a third wasfound within a predicted transmembrane domain.NADH dehydrogenase is the first and largest enzymecomplex in the respiratory chain of the oxidative phos-phorylation machinery, and plays a central role in energymetabolism [25,26]. A broad-scale study of adaptiveevolution of the mitochondrial genome of 41 placentalmammals revealed signatures of positive selection in theNADH dehydrogenase complex, largely restricted to theloop regions of the proton pumps, including ND5 [26].Additional studies [27-29] have also detected positivelyselected sites in ND5, with adaptive changes in thepiston arm suggested to have influenced fitness duringthe evolution of Pacific salmon species [29]. Futurestudies utilizing population level samples spanning theentire elevation gradient in BC are required to furtherinvestigate the role of ND5 in local adaptation of O.princeps across varying environments.Contigs unique to each elevationAdditional datasets were generated containing contigs thatwere only composed of transcriptome reads from either0 10 20 30 40 50 60 virion extracellular matrix synapse cell junction extracellular region membrane-enclosed lumen macromolecular complex membrane organelle cell electron carrier activity receptor activity structural molecule activity molecular transducer activity transporter activity enzyme regulator activity catalytic activity binding viral reproduction rhythmic process growth locomotion biological adhesion multi-organism process immune system process reproduction cell proliferation death signaling developmental process response to stimulus multicellular organismal process localization biological regulation metabolic process cellular process Cellular Component Molecular Function Biological Process Percent of genes Figure 2 Functional annotation of contigs in the high coverage dataset. The distribution of gene ontology (GO) terms is given for each ofeach of the three main GO categories (biological process, molecular function, and cellular component).Lemay et al. BMC Genomics 2013, 14:311 Page 5 of 11http://www.biomedcentral.com/1471-2164/14/311the high or low elevation (Table 2; Additional file 5).BLAST searches (BLASTx, NCBI nr database, maxe-value = 10-06) of these elevation-specific contigs produced88 positive matches in the high elevation dataset (meane-value = 1.8 × 10-08) and 83 positive hits among contigsunique to low elevation (mean e-value = 1.0 × 10-08).Interestingly, there was a high-elevation-specific contig(contig 31687; Additional file 5) that was a strong matchwith the hemoglobin alpha chain from high elevationsamples of both the Chinese red pika [O. erythrotis,GENEBANK: JX827174, e-value = 1.1 × 10-56] and the plat-eau pika [O. curzoniae, GENBANK: EF429202, e-value =1.2 × 10-55], species restricted to high elevation steppesin Asia (3000-5000m; [30]). An additional assembly ofraw reads to both the hemoglobin reference sequence[EF429202] and to the associated contig (contig 31687)confirmed that low elevation reads were indeed absent,rather than being misassembled during the initial denovo assembly and read-mapping (CLC GENOMICSWORKBENCH v.5.5, similarity 0.9, length fraction 0.5;data not shown).Hemoglobin is a key component of oxygen storage andregulation, and plays an important role in physiologicaladaptation to different environments [31]. A host ofstudies have demonstrated an association of hemoglobinalpha chain haplotype frequency with elevation in mammals[31-34]. Here, hemoglobin alpha chain transcripts wereonly detected among the high elevation sequencing reads.We Sanger sequenced the hemoglobin alpha chain in ourDNA samples of O. princeps, revealing no variation at thenucleotide level within or among elevations (data notshown). This result may be indicative of differential geneexpression across elevations, with expression among thelow elevation samples occurring below our detection level,even after the normalization of transcripts. Additional stud-ies are required to further elucidate the role of hemoglobinalpha chain, if any, in local adaptation of O. princeps.Gradients in latitude and elevation can been useful forpredicting the impact of climate change on naturalpopulations. For example, in the case of mountain species,low-elevation populations may possess unique geneticvariation associated with adaptation to higher temperature;if present, such adaptations might provide insight into theability of high elevation populations to adapt in responseto climate change. While our study was not designed totest predictions related to climate change, we providenovel sequence data from genes expressed by O. princepsat both low and high elevations, which provides a valuableresource for future research.ConclusionsTo our knowledge, this is the first broad-scale, transcrip-tome-level study conducted within the Ochotonidae,providing novel genomic resources to inform studies ofpika ecology, behaviour, and population history, whileenabling direct investigations of potential biotic responsesto changing environments. We identified 24,261 novelSNPs among O. princeps inhabiting different elevations.We detected SNPs and haplotypes that were fixed anddifferent among elevations, and identified the ND5 regionof the mitochondrial genome as a promising target genefor further studying the role of cellular respiration inlocal adaptation to varying environments. We also foundcontigs that were unique to each elevation, includinghemoglobin alpha chain, which may represent candidateregions subject to differential gene expression along thiselevation gradient. Although this RNAseq approach wassuccessful at identifying a large number of novel SNP loci,information on allele frequencies was limited by the smallnumber of individuals used in the pooled libraries. Emer-ging protocols that utilize combinatorial labelling methodsand Restriction Associated DNA (RAD; [35,36]) sequencingmay provide more efficient and cost-effective alternativesfor simultaneously discovering SNPs in non-model organ-isms and genotyping population-level samplings.MethodsSample collection, RNA extraction, and next generationsequencingSample collection was carried out in Tweedsmuir SouthProvincial Park in the Bella Coola Valley, BC, Canada,Table 3 Genetic diversity estimates from loci that weresuccessfully genotyped using HRM analysisLocus HO/HEBC high BC low Oregon high Oregon lowOcp687 0.33/0.28 0.25/0.47 0.00/0.00 0.00/0.00Ocp2098 0.00/0.00 0.50/0.50 0.00/0.00 0.00/0.00Ocp4162 0.29/0.49 0.00/0.00 0.00/0.00 0.00/0.50*Ocp4280 0.00/0.00 1.00/0.50* 0.22/0.35 0.33/0.44Ocp4649a 0.00/0.44 0.25/0.22 0.33/0.28 0.00/0.44Ocp5210 0.67/0.44 0.75/0.47 0.00/0.00 0.00/0.49*Ocp6361 0.00/0.00 0.00/0.38* 0.00/0.00 0.00/0.44Ocp6774 1.00/0.50 0.00/0.00 0.00/0.00 0.00/0.28Ocp7498 0.67/0.44 0.00/0.00 0.43/0.34 0.60/0.42Ocp8183 0.00/0.44 0.75/0.47 0.00/0.00 0.00/0.00Ocp8469 0.33/0.28 0.25/0.22 0.00/0.24 0.00/0.00Ocp14764 0.67/0.44 0.00/0.00 0.43/0.34 0.29/0.24Ocp15503 0.67/0.44 0.00/0.00 0.00/0.00 0.00/0.00Ocp15508 0.11/0.28 0.11/0.10 0.00/0.00 0.00/0.00Ocp17339 0.00/0.00 0.25/0.47 0.43/0.34 0.29/0.41Ocp102174 0.11/0.28 0.30/0.46 0.00/0.00 0.00/0.00Ocp102175 0.29/0.41 0.11/0.28 0.33/0.50 0.00/0.00* Significant deviation from HWE; bold denotes that the locus wasmonomorphic in that population.a X-linked locus.Lemay et al. BMC Genomics 2013, 14:311 Page 6 of 11http://www.biomedcentral.com/1471-2164/14/311which is a mountainous region with talus slopes scatteredthroughout. Previous work in Tweedsmuir Park hascharacterized neutral and adaptive genetic variation inO. princeps along three elevational transects [13]. Tissuecollection in the current study focussed on ‘The Hill’site, which has an elevational cline from 301 m (low eleva-tion site) to 1433m (high elevation site) above sea level. Arecent study demonstrated an average temperature dif-ference of up to six degrees between low and highelevation sites in summer [37], which is of a similarmagnitude to predicted temperature shifts for this partof the world during the next century. Three individualsat the low elevation site and three individuals at the highelevation site were collected using Tomahawk Live trapsand sacrificed in the field. Sample collection was carriedout in accordance with University of British ColumbiaAnimal Care Certificate #A07-0126 and sampling permitsfrom the BC Ministries of Environment (# 78470–25)and Forests, Lands and Natural Resource Operations(NA11-69259). Five tissue types (brain, gonad, heart,liver, lung) from each individual were immediatelyharvested and placed in separate 5 ml screw-cap vialscontaining 2.5 ml of RNALATER solution. Sampleswere held at 4°C for 24 hrs and then stored at −20°Cuntil needed. RNA was extracted from each tissueusing the RNEASY UNIVERSAL MINIKIT (Qiagen)following the manufacturer’s protocol. All specimens wereaccessioned within the mammal collection at the RoyalBritish Columbia Museum (RBCM catalogue numbers20919–20924; Additional file 6).                                10        20        30        40        50        60        70        80        90        100                                .         .         .         .         .         .         .         .         .         .O. princeps BC High MNLFSTLSALTILILTLPIFMSLTNFYLHPTFPTYVKNSVSLAFIISLVPTFIFLYTNQEIVLSNWHWTTIHTIKLSINLKLDFFSILFIPVALFVTWSIO. princeps BC Low MNLFSTLSALTILILTLPIFMSLTNFYLHPTFPTYVKNSVSLAFIISLVPTFIFLYTNQEIVLSNWHWTTIHTIKLSINLKLDFFSILFIPVALFVTWSIO. princeps NC005358 MNLFSTLSALTILILTLPIFMSLTNFYLHPTFPTYVKNSVSLAFIISLVPTVIFLYTNQEIVLSNWHWTTIHTIKFSINLKLDFFSILFIPVALFVTWSIO. cuniculus NC001913 MNLFSTSVAVSIIILVLPIVASFTNIFNSPNYPHYVKTSVSYAFTISLIPTLIFIATSQEMMVSNWHWMTIHTLKLTTSFKLDYFSMLFTPIALFVTWSI                                110       120       130       140       150       160       170       180       190       200  .         .         .         .         .         .         .         .         .         .O. princeps BC High IEFSLWYIHSDPNINRFFKYLLLFLITIIILVTANNLFQLFIGWEGVGIISFLLIGWWHGRTDANTAALQAILYNRIGDIGFVLSIAWFFIHINSWELQQO. princeps BC Low IEFSLWYIHSDPNINRFFKYLLLFLITIIILVTANNLFQLFIGWEGVGIISFLLIGWWHGRTDANTAALQAILYNRIGDIGFVLSIAWFFIHINSWELQQO. princeps NC005358 IEFSLWYIHSDPNINRFFKYLLLFLITIIILVTANNLFQLFIGWEGVGIISFLLIGWWHGRADANTAALQAILYNRIGDIGFVLSIAWFFIHINSWELQQO. cuniculus NC001913 MEFSMWYMHSDPKINQFFKYLLMFLITMLILVTANNMFQLFIGWEGVGIMSFLLIGWWHGRTDANTAALQAILYNRIGDIGFIMALAWFAINLNTWELQQ                                210       220       230       240       250       260       270       280       290       300                                .         .         .         .         .         .         .         .         .         .O. princeps BC High IFMLEQNNLTLPLIGLILAAAGKSAQFGLHPWLPTAIEGPTPVSALLHSSTIVVAGVFLLIRFYPILESNKLAQSLVLCLGALTTLFTALCALTQNDIKKO. princeps BC Low IFILEQNNLTLPLIGLILAAAGKSAQFGLHPWLPAAIEGPTPVSALLHSSTIVVAGVFLLIRFYPILESNKLAQSLVLCLGALTTLFTALCALTQNDIKKO. princeps NC005358 IFILEQNNLTLPLMGLILAAAGKSAQFGLHPWLPAAIEGPTPVSALLHSSTIVVAGVFLLIRFYPILESNKLAQSLVLCLGALTTLFTALCALTQNDIKKO. cuniculus NC001913 IFILDNNITILPLMGLILAATGKSAQFGLHPWLPSAMEGPTPVSALLHSSTMVVAGVFLLIRFYPLLENNKTAQTLILCLGAITTLFTALCALTQNDIKK                                310       320       330       340       350       360       370       380       390       400                                .         .         .         .         .         .         .         .         .         .O. princeps BC High IIAFSTSSQLGLIIVTIGINQPHLAFLHICTHAFFKAILFMCSGSIIHSLNDEQDIRKIGGLFNTLPFTSSALTIGSLALTGIPFLTGFYSKDLIIEAVNO. princeps BC Low IIAFSTSSQLGLIIVTIGINQPHLAFLHICTHAFFKAILFMCSGSIIHSLNDEQDIRKIGGLFNTLPFTSSALTIGSLALTGIPFLTGFYSKDLIIEAVNO. princeps NC005358 IIAFSTSSQLGLIIVTIGINQPHLAFLHICTHAFFKAMLFMCSGSIIHSLNDEQDIRKIGGLFNTLPFTSSALTIGSLALTGMPFLTGFYSKDLIIEAVNO. cuniculus NC001913 IVAFSTSSQLGLMMVTIGINQPHLAFLHICTHAFFKAMLFLCSGSIIHSLNDEQDIRKMGGLYKTMPFTASALTIGSLALTGMPFLTGFYSKDLIIESAN                                410       420       430       440       450       460       470       480       490       500  .         .         .         .         .         .         .         .         .         .O. princeps BC High TSYTNAWALLLTLIATSITAIYSTRVIFFALLNQPRFPPITTINENNPYLINSIKRLALGSIFAGFLISNNIPPFTVPPMTIPLYTKIAALTVTVLGFLLO. princeps BC Low TSYTNAWALLLTLIATSITAIYSTRVIFFALLNQPRFPPITTINENNPYLINSIKRLALGSIFTGFLISNNIPPFTVPPMTIPLYTKIAALTVTVLGFLLO. princeps NC005358 TSHTNAWALLLTLIATSITAIYSTRVIFFTLLNQPRFPPIVTINENNPYLTNSIKRLALGSIFAGFLISNNIPPFTVPPMTIPLYTKIAALMVTVLGFLLO. cuniculus NC001913 TSNTNAWALIITLIATSLTAVYSTRIIFFALLGQPRYPALIVINENNPLLINSIKRLALGSIFAGFLISNLITPNNVPQMTMPLYMKMTALFVTIMGFTI                                510       520       530       540       550       560       570       580       590       600                                .         .         .         .         .         .         .         .         .         .O. princeps BC High AIELNQLTLNLKLSPRSKLFYFSNLLGFFPTTIHRLIPYASLLFSLNTATTTLDITWTEKAIPKTISTIQINISSLISTQKGLIKLYSLSFLISITLAILO. princeps BC Low AIELNQLTLNLKLSPRSKLFYFSNLLGFFPTTIHRLIPYASLLFSLNTATTTLDITWTEKAIPKTISTIQINISSLISTQKGLIKLYSLSFLISITLAILO. princeps NC005358 AIELNQLTLNLKLNPHSKPFYFSNLLGFFPTTIHRLIPHASLLFSLNTATTTLDITWTEKAIPKTIATIQINLSSLISTQKGLIKLYSLTFLISITLAILO. cuniculus NC001913 AMELNQLSLSLKMTTQSPYFNFSNMLGFFPMTIHRILPYLNLSASQNMATLLLDMTWTEKAIPKNISDIQIFASTSVSSQKGLIKLYFLSFLISLLLVLFO. princeps BC High ILI*O. princeps BC Low ILI*O. princeps NC005358 ILI*O. cuniculus NC001913 ILT*Figure 3 Sequence alignment and secondary structure prediction of the ND5 gene in O. princeps haplotypes. The first two rows in thesequence alignment are from O. princeps sampled at high and low elevations in BC, respectively. The second two rows are the homologoussections from the published mitochondrial genomes of O. princeps [GENBANK: NC005358] and Oryctolagus cuniculus [GENBANK: NC001913].Amino acids in white bold and black background indicate non-synonymous substitutions fixed at low and high elevation pikas in BC. Predictedtransmembrane domains are shaded in gray. For the BC samples, residues 16–266 and 405–551 are the result of Sanger sequencing fourindividuals per elevation; the remaining residues are inferred from transcriptome read data of three individuals per elevation.Lemay et al. BMC Genomics 2013, 14:311 Page 7 of 11http://www.biomedcentral.com/1471-2164/14/311Two normalized cDNA libraries (Evrogen, Russia) wereconstructed using pooled RNA from all high elevation(5 tissues × 3 individuals) and low elevation (5 tissues × 3individuals) samples. The two resulting cDNA librarieswere each subject to a full run of 454 GS FLX Titaniumsequencing at the Genome Quebec core facility. Poolingof multiple individuals in each sample was used to providea preliminary indication of the genetic variation withinand among elevations; the combination of five tissue typesfor each individual was used to maximize the diversity ofexpressed genes present in each library. RNA sampleswere normalized in order to increase the detection of raretranscripts in the sequence data.AssemblyInitial trimming of the read data was performed usingthe CLC GENOMICS WORKBENCH (CLC Bio) v.4.8 suchthat very short reads (<100 bases), terminal nucleotides(five from each end), low quality reads (quality limit 0.05),and 454 sequencing adapters and primers were removedfrom the dataset. A de novo assembly using the CLCGENOMICS WORKBENCH v. 5.1 was then carried out(similarity = 0.90) in order to generate reference contigs.To facilitate a comparison of sequence variation betweenthe two elevations, the consensus sequence from eachreference contig was used to map the high and lowelevation reads separately (similarity = 0.90, length fraction0.5). We retained only those contigs that had a minimumlength of 200 bases and an average coverage greater than5× for each elevation (hereafter referred to as the highcoverage dataset).We performed a de novo assembly (CLC GENOMICSWORKBENCH v. 5.5; similarity = 0.90, length fraction0.5) using all contigs in the high coverage dataset in orderto identify contigs that partially overlapped. Redundanciesamong the contigs may be indicative of alternative splicingwithin the transcriptome data.We also generated datasets containing those contigsthat were composed of reads from only a single elevation(minimum coverage = 5×; minimum length = 200 bases).These two ‘elevation-unique’ datasets may suggest targetgenes for subsequent studies examining differences in geneexpression among elevations. For all analyses, assemblyand mapping parameters were optimized by comparingthe results of multiple runs at different levels of similarityand length fraction.Transcriptome annotationWe conducted sequence similarity searches for thehigh coverage dataset (n = 7,325 contigs) using Blast2GOv.2 [38,39]. For these analyses, a BLASTx search wasperformed using the NCBI nr database (maximum e-valuethreshold = 10-6, HSP length cut-off = 33, top 5 hits wereretained). In addition, gene ontology (GO) analysiswas carried out, which provides hierarchically structuredinformation with respect to molecular function, biologicalprocess, and cellular component. Annotations wereassigned using Blast2GO (maximum e-value threshold =10-6, HSP length cut-off = 20, GO weight 5). In addition,a BLAST search (BLASTx; same parameters as above)was carried out for all contigs in each of the twoelevation-unique datasets.SNP discoveryThe working dataset of high coverage contigs wasscreened for SNPs using the CLC GENOMICS WORK-BENCH v. 5.5 (minimum coverage 8×, minimum variantfrequency 10%, minimum number of reads per allele = 2,minimum central quality 20). SNP detection was carriedout separately for the two elevations and the resultingSNP tables were combined so that each site could becharacterized as either: (a) polymorphic in both eleva-tions; (b) fixed in one elevation, polymorphic in theother; or (c) fixed for different alleles in each elevation.In order to putatively identify the sites of greatest differ-entiation between elevations, a divergence value basedon the index implemented in Juekens et al. [40] wascalculated for each SNP, defined as the absolute valueof the difference in the frequency of the major alleleamong elevations.SNP validationA panel of SNPs with divergence values ≥50% wasused to genotype an independent sample of O. princeps inorder to test this ascertainment procedure. Validation ofcandidate SNPs was carried out following a pipelinesimilar to that implemented by Seeb et al. [23]. Briefly,primers were designed using PRIMER3 [41] such thatthey would amplify an ~200bp fragment that encompasseda single SNP (Additional file 4). An initial PCR wasused to identify loci that produced a single cleanproduct of the anticipated size; these loci were then usedto genotype 42 individuals using High Resolution Melt(HRM) analysis (see below).Each test PCR contained 1.25 μl of 10× buffer, 1.25 μlof 2 mM dNTP mix (Kapa Biosystems), 1.0 μl of BSA,0.5 μl of 10 mM forward and reverse primer, 0.5 units ofTaq polymerase (AmpliTaq Gold, Applied Biosystems),20–100 ng of DNA template, and ultra pure water for atotal reaction volume of 12.5 μl. For each reaction, atouchdown PCR procedure was implemented using aVeriti thermal cycler (Applied Biosystems). The programhad an initial denaturation at 95°C for 10 minutes, followedby 8 cycles at 95°C for 30 seconds, 59°C for 30 seconds,and 72°C for 30 seconds with the annealing temperaturedecreasing by 1.0°C per cycle. This was followed by 27cycles at 94°C for 30 seconds, 51°C for 30 seconds, and72°C for 30 seconds. The final cycle had an extension ofLemay et al. BMC Genomics 2013, 14:311 Page 8 of 11http://www.biomedcentral.com/1471-2164/14/31172°C for 10 minutes and was then held at 4°C. PCRproducts were run on a 1.5% agarose gel in order to ob-tain a preliminary assessment of the quality and size of theamplicon. Loci that failed to amplify, showed evidence forthe presence of introns (larger products than expected), orhad multiple bands were not retained for subsequentanalyses.High resolution melt analysis was carried out usingDNA samples of O. princeps from both the Bella CoolaValley, BC, Canada and the Columbia River Gorge, Oregon,USA. Sampling procedures for the Bella Coola sampleshave been previously reported by Henry et al. [13]. Weused DNA samples from the same high elevation (n = 10individuals) and low elevation (n = 11 individuals) sitesfrom which the tissue samples for the transcriptomesequencing were collected. Oregon samples were collectedin the summer of 2012 from sites at both high (n = 11) andlow elevations (n = 10) using non-invasive hair-snares [42].DNA extraction was carried out using a DNA IQ™ Tissueand Hair Extraction Kit (Promega, Madison, WI, USA) kitfollowing the protocol outlined in Henry et al. [13].Each HRM reaction contained 7.2 μl of Precision MeltSupermix (BioRad), 0.4 μl of each primer, 20–100 ng ofDNA template, and ultra pure water for a total reactionvolume of 20 μl. High resolution melt analyses were runin 96 well plates on a BioRad CFX96 Touch™ real timePCR detection system. A two-step touchdown PCRprotocol was used, starting with an initial denaturationstep at 95°C for 2 minutes, followed by 9 cycles of 95°Cfor 10 seconds, 60°C for 30 seconds, with the annealingtemperature decreasing by 1°C per cycle. This wasfollowed by 43 cycles of 95°C for 10 seconds and 50°Cfor 30 seconds. The final PCR cycle consisted of 95°Cfor 30 seconds followed by 55°C for 1 minute. A plateread was obtained at the end of every PCR cycle. Themelt curve data were collected starting at 70°C andincreasing by 0.2°C every 10 seconds to a maximumof 95°C. A plate read was obtained at every 0.2°C increment.Melt curve data were analyzed using BioRad PrecisionMelt Analysis™ software.Loci that successfully amplified, were free of introns, andproduced well-resolved clusters of HRM curves (n = 26loci) were subjected to Sanger re-sequencing on an ABI3130XL Genetic Analyzer (Applied Biosystems). Giventhat each HRM cluster should represent a single SNPgenotype, 2–3 individuals from each cluster weresequenced in order to determine the genotype of eachcluster. For each locus that was successfully genotyped,we calculated the expected and observed heterozygosityvalues, tested for linkage disequilibrium, and testedfor deviations from Hardy-Weinberg equilibrium usingGENEPOP v.4 [43]. Type I error rates were corrected formultiple comparisons using the sequential Bonferroniprocedure [44].Mitochondrial DNAMitochondrial genes may represent an important compo-nent of adaptation to different elevations in O. princeps[45]. To assess the prevalence and sequence variation ofmitochondrial genes in the transcriptome data, we usedthe annotated mitochondrial genome for O. princeps[GENBANK: NC005358] as a reference for read mapping(CLC GENOMICS WORKBENCH v.5.5, similarity = 0.90,length fraction 0.5).For ND5, Sanger sequences were obtained from fourhigh and four low elevation pikas from BC using primersdesigned based on transcriptome sequence data fromcontig 1829 and contig 24554 (Additional file 4). A touch-down PCR protocol was used, with an initial denaturationat 95°C for 10 minutes, then 8 cycles at 95°C for 30seconds, 59°C for 30 seconds, and 72°C for 2 minutes.This was followed by 32 cycles at 95°C for 30 seconds, 51°Cfor 30 seconds, and 72°C for 2 minutes. The final cyclehad an extension of 72°C for 7 minutes and was then heldat 4°C. PCR products were purified using Exo-Sap-It (USBCorporation) and sequenced on an ABI 3130XL GeneticAnalyzer (Applied Biosystems). The resulting data werealigned with previously published ND5 sequences from O.princeps [GENBANK: NC005358] and the Europeanrabbit [Oryctolagus cuniculus; GENBANK: NC001913].Transmembrane helices were predicted from translatedamino acid sequences using the hidden Markov modelimplemented in TMHMM v2.0 [46].Additional filesAdditional file 1: High coverage contig sequences. Text file (.txt) inFASTA format containing the sequence of all high coverage contigs usedfor SNP detection (minimum length = 200 bases, minimum coverage = 5×for each ecotype).Additional file 2: Characterization of high coverage contigs. Excelfile (.csv) listing the length, coverage, number of reads, and top BLASTxhit for each of the high coverage contigs.Additional file 3: SNP information. Excel file (.csv) characterizing the24,261 SNPs identified in the high coverage dataset.Additional file 4: SNP primers. A table (.doc) containing the primersequences for all loci for which HRM validation was attempted.Additional file 5: Elevation unique contig sequences. Text file (.txt) inFASTA format containing the sequence of all contigs that were composedentirely of reads from a single elevation (minimum length = 200 bases,minimum coverage = 5×).Additional file 6 Sample collection. Word document (.doc) describingthe samples used to generate each cDNA library.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsMR and ML designed the study. ML extracted the RNA, analyzed the data,and prepared the manuscript. PH collected samples, assisted with dataanalysis and helped draft the manuscript. CL and KR collected samples, andassisted with genotyping and Sanger sequencing. MR obtained funding,collected samples, analyzed the data and helped to draft the manuscript. Allauthors have read and approved the final manuscript.Lemay et al. BMC Genomics 2013, 14:311 Page 9 of 11http://www.biomedcentral.com/1471-2164/14/311AcknowledgementsWe thank Joanna Varner, Brody Granger, Daniel Rissling, Alison Henry, ZijianSim and Megan Perra for field assistance. We also thank Mark Rheault forproviding access to some necessary laboratory equipment. The pikaphotograph was provided by Alison Henry. This work was funded by aGenome BC Strategic Opportunities Fund grant # 130 (MR) and NSERCDiscovery Grant # 341711–07 (MR). ML was partially supported by an NSERCPostgraduate Scholarship. PH was partially supported by a grant from theSwiss National Science Foundation. CL was partially supported by anUndergraduate Research Award from the Irving K. Barber School of Arts andSciences at UBC’s Okanagan campus.Author details1Department of Biology, University of British Columbia, Okanagan Campus3333 University Way, Kelowna, BC V1V 1V7, Canada. 2Present address:Ecosystem Science and Management Program, University of Northern BritishColumbia, 3333 University Way, Prince George, BC V2N 4Z9, Canada.Received: 5 December 2012 Accepted: 4 May 2013Published: 10 May 2013References1. Parmesan C: Ecological and evolutionary responses to recent climatechange. Annu Rev Ecol, Evol Syst 2006, 37:637–669.2. Sgro CM, Lowe AJ, Hoffmann AA: Building evolutionary resilience forconserving biodiversity under climate change. Evol Appl 2011, 4:326–337.3. Beever EA, Ray C, Mote PW, Wilkening JL: Testing alternative models ofclimate-mediated extirpations. Ecol Appl 2010, 20:164–178.4. Reusch TBH, Wood TE: Molecular ecology of global change. Mol Ecol 2007,16:3973–3992.5. Galbreath KE, Hafner DJ, Zamudio KR: When cold is better: Climate-drivenelevation shifts yield complex patterns of diversification anddemography in an alpine specialist (the American pika, Ochotonaprinceps). Evolution 2009, 63:2848–2863.6. Smith AT, Weston ML: Ochotona princeps. Mammalian Species 1990,352:1–8.7. Jeffress MR, Rodhouse TJ, Ray C, Wolff S, Epps C: The idiosyncrasies ofplace: geographic variation in the climate-distribution relationships ofthe American pika. Ecol Appl. In press.8. Mead J: Quaternary records of pika, Ochotona, in North America. Boreas1987, 16:165–171.9. Beever EA, Brussard PE, Berger J: Patterns of apparent extirpation amongisolated populations of pikas (Ochotona princeps) in the Great Basin.J Mammal 2003, 84:37–54.10. Beever EA, Smith AT: Ochotona princeps. IUCN Red List of Threatened Species,version 2012.2. www.iucnredlist.org/details/41267/0.11. Peacock MM, Smith AT: Nonrandom mating in pikas Ochotona princeps:evidence for inbreeding between individuals of intermediaterelatedness. Mol Ecol 1997, 6:801–811.12. Calkins MT, Beever EA, Boykin KG, Frey JK, Andersen MC: Not-so-splendidisolation: modeling climate-mediated range collapse of a montanemammal Ochotona princeps across numerous ecoregions. Ecography2012, 35:780–791.13. Henry P, Sim ZJ, Russello MA: Genetic Evidence for Restricted Dispersalalong Continuous Altitudinal Gradients in a Climate Change-SensitiveMammal: The American Pika. PLoS One 2012, 7:e39077.14. Smith AT: Distribution and dispersal of pikas: Influences of behavior andclimate. Ecology 1974, 55:1368–1376.15. Beever EA, Ray C, Wilkening JL, Brussard PF, Mote PW: Contemporaryclimate change alters the pace and drivers of extinction. Global ChangeBiol 2011, 17:2054–2070.16. Smith AT, Weidong L, Hik DS: Pikas as harbingers of global warming.Species 2004, 41:4–5.17. Narum SR, Hess JE: Comparison of F-ST outlier tests for SNP loci underselection. Mol Ecol Resour 2011, 11:184–194.18. Luikart G, England PR, Tallmon D, Jordan S, Taberlet P: The power andpromise of population genomics: from genotyping to genome typing.Nat Rev Genet 2003, 4:981–994.19. Garvin MR, Saitoh K, Gharrett AJ: Application of single nucleotidepolymorphisms to non-model species: a technical review. Mol Ecol Resour2010, 10:915–934.20. Brumfield RT, Beerli P, Nickerson DA, Edwards SV: The utility of singlenucleotide polymorphisms in inferences of population history. TrendsEcol Evol 2003, 18:249–256.21. Willing EM, Bentzen P, van Oosterhout C, Hoffmann M, Cable J, Breden F,Weigel D, Dreyer C: Genome-wide single nucleotide polymorphismsreveal population history and adaptive divergence in wild guppies. MolEcol 2010, 19:968–984.22. Flicek P, Amode MR, Barrell D, Beal K, Brent S, Carvalho-Silva D, Clapham P,Coates G, Fairley S, Fitzgerald S, Gil L, Gordon L, Hendrix M, Hourlier T, JohnsonN, Kaehaeri AK, Keefe D, Keenan S, Kinsella R, Komorowska M, Koscielny G,Kulesha E, Larsson P, Longden I, McLaren W, Muffato M, Overduin B, PignatelliM, Pritchard B, Riat HS, et al: Ensembl 2012. Nucleic Acids Res 2012, 40:D84–D90.23. Seeb JE, Pascal CE, Grau ED, Seeb LW, Templin WD, Harkins T, Roberts SB:Transcriptome sequencing and high-resolution melt analysis advancesingle nucleotide polymorphism discovery in duplicated salmonids. MolEcol Resour 2011, 11:335–348.24. Liew M, Pryor R, Palais R, Meadows C, Erali M, Lyon E, Wittwer C:Genotyping of single-nucleotide polymorphisms by high-resolutionmelting of small amplicons. Clin Chem 2004, 50:1156–1164.25. Brandt U: Energy converting NADH : Quinone oxidoreductase (ComplexI). Annu Rev Biochem 2006, 75:69–92.26. da Fonseca RR, Johnson WE, O'Brien SJ, Ramos MJ, Antunes A: Theadaptive evolution of the mammalian mitochondrial genome. BMCGenomics 2008, 9:119.27. Pabijan M, Spolsky C, Uzzell T, Szymura JM: Comparative analysis ofmitochondrial genomes in Bombina (Anura; Bombinatoridae). J Mol Evol2008, 67:246–256.28. Mishmar D, Ruiz-Pesini E, Mondragon-Palomino M, Procaccio V, Gaut B,Wallace DC: Adaptive selection of mitochondrial complex I subunitsduring primate radiation. Gene 2006, 378:11–18.29. Garvin MR, Bielawski JP, Gharrett AJ: Positive Darwinian Selection in thePiston That Powers Proton Pumps in Complex I of the Mitochondria ofPacific Salmon. PLoS One 2011, 6:e24127.30. Yingzhong Y, Yue C, Guoen J, Zhenzhong B, Lan M, Haixia Y, Rili G:Molecular cloning and characterization of hemoglobin alpha and betachains from plateau pika (Ochotona curzoniae) living at high altitude.Gene 2007, 403:118–124.31. Storz JF: Hemoglobin function and physiological adaptation to hypoxiain high-altitude mammals. J Mammal 2007, 88:24–31.32. Storz JF, Runck AM, Sabatino SJ, Kelly JK, Ferrand N, Moriyama H, Weber RE,Fago A: Evolutionary and functional insights into the mechanismunderlying high-altitude adaptation of deer mouse hemoglobin. ProcNatl Acad Sci U.S.A. 2009, 106:14450–14455.33. Monge C, Leonvelarde F: Physiological adaptation to high altitude:oxygen transport in mammals and birds. Physiol Rev 1991, 71:1135–1172.34. Weber RE, Ostojic H, Fago A, Dewilde S, Van Hauwaert ML, Moens L, MongeC: Novel mechanism for high-altitude adaptation in hemoglobin of theAndean frog Telmatobius peruvianus. Am J Physiol Regul Integr CompPhysiol 2002, 283:R1052–R1060.35. Peterson BK, Weber JN, Kay EH, Fisher HS, Hoekstra HE: Double DigestRADseq: An Inexpensive Method for De Novo SNP Discovery andGenotyping in Model and Non-Model Species. PLoS One 2012, 7:e37135.36. Baird NA, Etter PD, Atwood TS, Currey MC, Shiver AL, Lewis ZA, Selker EU,Cresko WA, Johnson EA: Rapid SNP Discovery and Genetic MappingUsing Sequenced RAD Markers. PLoS One 2008, 3:e3376.37. Henry P, Henry A, Russello MA: Variation in habitat characteristics ofAmerican pikas along an elevation gradient at their northern rangemargin. Northwest Sci 2012, 86:346–250.38. Conesa A, Gotz S, Garcia-Gomez JM, Terol J, Talon M, Robles M: Blast2GO: auniversal tool for annotation, visualization and analysis in functionalgenomics research. Bioinformatics 2005, 21:3674–3676.39. Gotz S, Garcia-Gomez JM, Terol J, Williams TD, Nagaraj SH, Nueda MJ, RoblesM, Talon M, Dopazo J, Conesa A: High-throughput functional annotation anddata mining with the Blast2GO suite. Nucleic Acids Res 2008, 36:3420–3435.40. Jeukens J, Renaut S, St-Cyr J, Nolte AW, Bernatchez L: The transcriptomicsof sympatric dwarf and normal lake whitefish (Coregonus clupeaformisspp., Salmonidae) divergence as revealed by next-generationsequencing. Mol Ecol 2010, 19:5389–5403.41. Rozen S, Skaletsky HJ: Primer3 on the WWW for general users and for biologistprogrammers. In Bioinformatics Methods and Protocols: Methods for MolecularBiology. Edited by Krawetz S, Misener S. Totowa, NJ: Humana Press; 2000:365–386.Lemay et al. BMC Genomics 2013, 14:311 Page 10 of 11http://www.biomedcentral.com/1471-2164/14/31142. Henry P, Russello MA: Obtaining high-quality DNA from elusive smallmammals using low-tech hair snares. Eur J Wildlife Res 2011, 57:429–435.43. Raymond M, Rousset F: GENEPOP (version-1.2) – population geneticssoftware for exact tests and ecumenicism. J Hered 1995, 86:248–249.44. Rice WR: Analyzing tables of statistical tests. Evolution 1989, 43:223–225.45. Luo YJ, Gao WX, Gao YQ, Tang S, Huang QY, Tan XL, Chen J, Huang TS:Mitochondrial genome analysis of Ochotona curzoniae and implicationof cytochrome c oxidase in hypoxic adaptation. Mitochondrion 2008,8:352–357.46. Krogh A, Larsson B, von Heijne G, Sonnhammer ELL: Predictingtransmembrane protein topology with a hidden Markov model:Application to complete genomes. J Mol Biol 2001, 305:567–580.doi:10.1186/1471-2164-14-311Cite this article as: Lemay et al.: Novel genomic resources for a climatechange sensitive mammal: characterization of the American pikatranscriptome. BMC Genomics 2013 14:311.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitLemay et al. BMC Genomics 2013, 14:311 Page 11 of 11http://www.biomedcentral.com/1471-2164/14/311


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items