Quantitative and multiplexed DNA methylation analysis using long-read single-molecule real-time bisulfite sequencing (SMRT-BS)
© Yang et al.; licensee BioMed Central. 2015
Received: 17 November 2014
Accepted: 23 April 2015
Published: 6 May 2015
DNA methylation has essential roles in transcriptional regulation, imprinting, X chromosome inactivation and other cellular processes, and aberrant CpG methylation is directly involved in the pathogenesis of human imprinting disorders and many cancers. To address the need for a quantitative and highly multiplexed bisulfite sequencing method with long read lengths for targeted CpG methylation analysis, we developed single-molecule real-time bisulfite sequencing (SMRT-BS).
Optimized bisulfite conversion and PCR conditions enabled the amplification of DNA fragments up to ~1.5 kb, and subjecting overlapping 625–1491 bp amplicons to SMRT-BS indicated high reproducibility across all amplicon lengths (r = 0.972) and low standard deviations (≤0.10) between individual CpG sites sequenced in triplicate. Higher variability in CpG methylation quantitation was correlated with reduced sequencing depth, particularly for intermediately methylated regions. SMRT-BS was validated by orthogonal bisulfite-based microarray (r = 0.906; 42 CpG sites) and second generation sequencing (r = 0.933; 174 CpG sites); however, longer SMRT-BS amplicons (>1.0 kb) had reduced, but very acceptable, correlation with both orthogonal methods (r = 0.836-0.897 and r = 0.892-0.927, respectively) compared to amplicons less than ~1.0 kb (r = 0.940-0.951 and r = 0.948-0.963, respectively). Multiplexing utility was assessed by simultaneously subjecting four distinct CpG island amplicons (702–866 bp; 325 CpGs) and 30 hematological malignancy cell lines to SMRT-BS (average depth of 110X), which identified a spectrum of highly quantitative methylation levels across all interrogated CpG sites and cell lines.
SMRT-BS is a novel, accurate and cost-effective targeted CpG methylation method that is amenable to a high degree of multiplexing with minimal clonal PCR artifacts. Increased sequencing depth is necessary when interrogating longer amplicons (>1.0 kb) and the previously reported bisulfite sequencing PCR bias towards unmethylated DNA should be considered when measuring intermediately methylated regions. Coupled with an optimized bisulfite PCR protocol, SMRT-BS is capable of interrogating ~1.5 kb amplicons, which theoretically can cover ~91% of CpG islands in the human genome.
KeywordsDNA methylation CpG islands Bisulfite sequencing Long-read sequencing Third generation sequencing Single-molecule real-time (SMRT) sequencing Pacific Bioscience
DNA methylation plays an essential role in several critical cellular processes including gene expression regulation, imprinting and X chromosome inactivation, and deregulation of the epigenetic machinery has been directly implicated in both Mendelian disorders and tumorigenesis [1-4]. In mammals, DNA methylation predominantly occurs at the 5-position of the cytosine pyrimidine ring within a CpG dinucleotide. Many techniques to detect CpG methylation have been developed that can be broadly classified by underlying chemistry, including bisulfite conversion, methylation sensitive restriction enzyme analyses, and methylated DNA immunoprecipitation [5-9]. Each has its own advantages and disadvantages; however, bisulfite conversion has been the most widely used sample preparation method for DNA methylation analysis. Common methylation detection techniques coupled with bisulfite conversion include PCR-based assays, microarrays, sequencing and other approaches, which differ in their capacities for single-nucleotide resolution, quantitation, and throughput.
Since the initial 1992 report on CpG methylation detection using bisulfite Sanger sequencing , the technique has been widely used for methylation discovery and as a diagnostic assay for detecting methylation abnormalities at imprinting control regions and the CpG islands of specific genes [11,12]. Importantly, the ongoing advances in sequencing technology have resulted in additional platforms available to execute bisulfite sequencing, including pyrosequencing [13,14] and next-generation sequencing [15-17]. Although laborious and low-throughput, bisulfite Sanger sequencing historically has been one of the most accurate and quantitative techniques to assess CpG methylation. Coupled with PCR amplicon cloning, bisulfite Sanger sequencing allows for allele-specific CpG methylation assessment; however, this is offset by a costly and time-consuming protocol and an inability to multiplex. Bisulfite pyrosequencing offers a faster, reproducible and quantitative analysis of DNA methylation, but is restricted to short read lengths (~150 bp) with limited capacity for multiplexing . Although a serial pyrosequencing technique using multiple primers has been developed for extended read lengths, it is still constrained by short PCR amplicon lengths . Additionally, it should be noted that all standard bisulfite conversion-based CpG methylation detection methods cannot discriminate 5-hydroxymethylcytosine from 5-methylcytosine. Consequently, oxidative bisulfite sequencing  and microarray  methods have recently been developed to distinguish these related nucleotides.
In contrast to targeted bisulfite sequencing techniques, next-generation bisulfite sequencing, including reduced representation bisulfite sequencing (RRBS)  and whole-genome bisulfite sequencing , can profile DNA methylation across an entire genome by quantitatively assessing the majority of CpG sites in a single experiment. This unparalleled resolution in methylation detection is transforming the field of epigenomics; however, adopting these genome-wide approaches requires significant bioinformatics expertise and may be prohibitively expensive for certain applications given their low throughput. An example of multiplexed and targeted bisulfite sequencing using 454 next-generation sequencing chemistry has been reported; however, the utility of this highly quantitative method was restricted to read lengths of only ~130 bp .
To address the need for a quantitative and highly multiplexed targeted bisulfite sequencing method with long read lengths, we developed a technique that combines bisulfite conversion with third-generation single-molecule real-time (SMRT) sequencing. Coupled with an optimized long-range bisulfite amplification protocol and empowered by the long read lengths of SMRT sequencing (up to ~20 kb) , multiplexed SMRT bisulfite sequencing (SMRT-BS) can accurately measure CpG methylation across ~1.5 kb regions without the need for PCR amplicon subcloning. As a cost-effective alternative to other targeted bisulfite sequencing techniques, SMRT-BS is an efficient and highly quantitative method for DNA methylation analysis.
Results and discussion
SMRT Bisulfite Sequencing (SMRT-BS) procedure
Since bisulfite conversion of DNA results in a substantial increase in adenine (A) and thymine (T) content and previous reports have indicated that A/T rich DNA can be more efficiently amplified by reduced PCR extension temperatures , long-range bisulfite PCR was evaluated using a gradient of extension temperatures. Treated DNA from each of the bisulfite conversion kits were subjected to PCR with two amplicon lengths (655 and 1109 bp) and extension temperatures of 65°C, 68°C, and 72°C (Figure 2B). These data indicated that two bisulfite conversion kits (Epigentek Methylamp and Qiagen EpiTect) and a PCR extension temperature of 65°C resulted in the most consistent amplification of the longer 1109 bp product using the reported conditions.
Long-range amplification and SMRT-BS quality filtering
To determine the maximum PCR amplicon size achievable with bisulfite converted DNA from the Qiagen and Epigentek kits, eight primer sets were tested that ranged in amplicon length from 655–4027 bp (Figure 2C, Additional file 1: Table S1). All primers were previously validated by short-length PCR to ensure their capacity for amplification. These results indicated that bisulfite PCR with DNA converted by the Epigentek Methylamp kit using the reported conditions could amplify products up to ~2.0 kb (Figure 2D), although amplification of the 2.0 kb product was not generally consistent. The ~1.5 kb product was consistently amplified and was determined to be the approximated upper limit of stable bisulfite PCR using these conditions.
Notably, the ability to amplify longer amplicons following bisulfite conversion with the Epigentek Methylamp kit was not reflective of poor bisulfite conversion efficiency as the cytosine conversion rate of amplicons subjected to SMRT-BS (detailed below) had an average conversion rate of 97.0%. All amplicons with conversion rates less than 95% (6.8% of all reads) were filtered prior to CpG methylation quantitation, which increased the average conversion rate to 97.3%. In addition, potential clonal PCR artifacts, as defined by identical patterns of both CpG and non-CpG cytosines throughout the amplicons (0.3% of all reads) , were filtered prior to CpG methylation quantitation.
Long-read SMRT-BS and reproducibility
In addition to assessing the correlation between overall amplicon methylation, individual CpG site methylation levels were compared and the average standard deviations (SDs) between triplicates in each tested amplicon are illustrated in Figure 3D. Of note, the median SDs for all CpG sites across all triplicate amplicons was less than ~0.10. The unmethylated and highly methylated CpG islands were very consistent at all amplicon lengths and replicates (median SD = 0.009, 0.009, 0.023 and 0.016 for the unmethylated amplicons, and SD = 0.010, 0.012, 0.032 and 0.024 for the highly methylated amplicons) (Figure 3D). However, analysis of the CpG island with an intermediate methylation level using the four different sized amplicons identified CpG site methylation median SDs of 0.044, 0.040, 0.097 and 0.111, respectively.
As also illustrated in Figure 3d, the amplicons that were greater than ~1 kb had lower SMRT-BS sequencing depth than the shorter amplicons despite being pooled with equal numbers of molecules. This resulted in methylation reproducibility being positively correlated with sequencing depth and negatively correlated with amplicon length. Consequently, these data underscore the need for increased sequencing depth for regions with intermediate methylation levels compared to those with low or high methylation levels to achieve similar confidence intervals and margins of error. Shorter amplicons may get preferentially sequenced when pooled with longer amplicons, which could be counterbalanced by increasing the molecule quantity of longer amplicons in the SMRT sequencing reaction when multiplexed with shorter amplicons. In addition, the increased variability observed in the intermediate methylation amplicons also suggests that PCR bias towards unmethylated DNA may occur, as has previously been reported with traditional bisulfite sequencing .
Validation of SMRT-BS methylation quantitation
SMRT-BS multiplexing and utility
Many techniques have been developed to measure CpG methylation, which differ widely in their chemistry, sample requirement, resolution, capacity for quantitation, throughput, accessibility and cost (for recent review, see ). Although genome-wide sequencing approaches are often desirable, the immediate benefits in CpG coverage can be offset by high per sample costs, a low throughput, and the need for computational expertise. Consequently, targeted approaches to detect CpG methylation are still widely used and continually developed, now often coupling bisulfite conversion with next-generation sequencing [15-17]. However, the paucity of targeted bisulfite sequencing techniques capable of examining amplicons longer than the typical bisulfite PCR size range (~300-500 bp) [15-17,25-27] prompted our development of SMRT-BS.
Although SMRT sequencing is capable of directly detecting modified nucleotides by their unique polymerase kinetics without bisulfite conversion , the introduction of an amplification step abolishes its ability to detect any epigenetic modifications of template DNA. A key component to the development of SMRT-BS was the optimization of bisulfite conversion and PCR, which resulted in amplicons up to ~1.5-2.0 kb from bisulfite-converted DNA. Subjecting these amplicons to multiplexed SMRT-BS indicated that the technique was reproducible and highly concordant with other lower throughput quantitative CpG methylation methods. However, the increased variability observed with SMRT-BS of intermediately methylated regions must be acknowledged as a potential limitation that may be mitigated by increased sequencing depth. Similarly, as has previously been reported for amplicon bisulfite sequencing , it should be noted that SMRT-BS of intermediately methylated regions may also be influenced by PCR bias towards unmethylated DNA. To account for this known problem inherent to amplicon bisulfite sequencing, calibration strategies using standard curves of methylation controls have been reported if needed [30,33].
As a targeted amplicon sequencing method and an extension of previous bisulfite sequencing techniques, SMRT-BS is capable of a high degree of multiplexing and sequencing depth for accurate methylation quantitation with negligable clonal PCR artifacts. Importantly, this technique also takes advantage of the long read lengths achievable with SMRT sequencing, which allows for more thorough regional CpG methylation assessment and increases the capacity for studying the relationship between phased single nucleotide variants and allele-specific CpG methylation. Additional potential applications of SMRT-BS include targeted panels for diagnostic CpG methylation detection, and multiplexed interrogation of candidate CpG methylation sites for epigenome-wide association study (EWAS) [34,35] replication analyses. Notably, the size distribution of all CpG islands in the hg19 human genome reference sequence (as defined by the UCSC Genome Browser and reference ) suggests that the ability of SMRT-BS to measure DNA methylation across ~1.5 kb amplicons could theoretically interrogate ~91% of all the CpG islands in the human genome.
Specimens and cell lines
Peripheral blood samples from healthy donors who self-reported their racial background and gave informed consent for the use of their DNA for research were obtained from the New York Blood Center with IRB approval as previously defined . All personal identifiers were removed, and isolated DNA samples were tested anonymously. Genomic DNA was isolated using the Puregene® DNA Purification kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions.
The human acute myeloid leukemia (AML193, THP-1, KG1A, MV4-11, KCL-22, OCI-M1), chronic myeloid leukemia (LAMA-84, JURL-MK1, K562, KU-812), anaplastic large cell lymphoma (SU-DHL-1, SUP-M2, SR786, DEL), plasma cell leukemia (SK-MM-2, L-363, AMO-1, JJN), Burkitts lymphoma (BL-41, BL-70, CA-46, RAJI), B–cell lymphoma (DOHH, OCI-LY-2, OCI-LY-3, OCI-LY-7, OCI-LY-8, OCI-LY-10, OCI-LY-19) and multiple myeloma (KMS-12-BM) cell lines were obtained from either the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany) or the American Type Culture Collection (ATCC; Manassas, VA, USA). All cell lines were cultured as per recommended conditions. Genomic DNAs were isolated using the DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions.
All bisulfite treatments of genomic DNAs (1 μg) were performed according to the manufacturer’s instructions. Six commercial kits were evaluated: (1) Diagenode Premium Bisulfite Kit, (2) Epigentek Methylamp DNA Modification Kit, (3) NEB EpiMark Bisulfite Conversion Kit, (4) Promega MethylEdge Bisulfite Conversion System, (5) Qiagen EpiTect Bisulfite Kit, and (6) Zymo EZ-DNA Methylation-Gold Kit. Bisulfite-treated DNAs were quantified using the NanoDrop 1000 and sized with a 2100 Bioanalyzer using the RNA6000 Pico kit (Agilent Technologies).
First-step PCR reactions were performed in 20 μl containing ~50-100 ng of bisulfite-treated DNA, 1X PCR Buffer, 0.2 mM of each dNTP, 0.2 μM forward and reverse region-specific primers (Additional file 1: Tables S1-S3), and 1.0 unit of TaKaRa Taq HS (Clontech Laboratories, Inc., Mountain View, CA, USA). Amplification consisted of an initial denaturation step at 94°C for 2 min followed by 35 amplification cycles (94°C for 20 sec, 55°C for 45 sec, and 65°C for 1 min/kb + 30 seconds) and a final incubation at 65°C for 5 min (unless otherwise noted). Second-step PCR reactions were carried out using 1 μl of a 1:50 dilution of first-step PCR product in a total volume of 25 μl with the same conditions as first-step PCR, but using barcoded universal primers and an annealing temperature of 60°C.
Single-Molecule Real-Time (SMRT) sequencing
Where V i is the volume of each PCR amplicon, M is the total mass of pooled PCR amplicons, L i is the length of each amplicon, n is the total number of samples, C i is the concentration of each amplicon and m is the total number of amplicons. A total of 500 ng of pooled PCR amplicons were submitted for SMRT sequencing.
Single-molecule real-time (SMRT) sequencing was performed according to the P5-C3 Pacific Biosciences protocol with a movie collection time of 180 minutes. In brief, pooled PCR amplicons were quantified using Qubit fluorometric analysis (Life Technologies) and a Bioanalysis 12000 chip (Agilent Technologies) to assess PCR amplicon quality, size, and quantity. Additionally, barcoded and pooled amplicons were purified using Ampure XP Solid Phase Reversible Immobilization (Beckman Coulter) at 0.8-fold volume. SMRTbell libraries were constructed using end-repair, ligation, and exonuclease purification strategies detailed in the Pacific Biosciences P5-C3 Template Preparation Kit protocols. SMRTbell templates were then bound to polymerase molecules for 4 hours at 25°C using 3 nM of the amplicon SMRTbell library and excess P5 DNA polymerase at a concentration of 9 nM as previously described . The polymerase-template complexes were immobilized at 250 pM for 30 min on nanofabricated SMRTcells containing an array of zero-mode waveguides (ZMWs), and ZMWs were analyzed for sequencing to generate reads using a 1x180-minute collection protocol. Circular consensus sequencing (CCS) was employed using multiple passes on each SMRTbell to generate CCS reads with higher accuracy for data analysis. The ‘Reads of Insert’ pipeline was utilized with a filter of 85% accuracy and 1-pass prior to downstream analyses.
SMRT-BS data analysis
Infinium HumanMethylation450 BeadChip (450K-array)
Methylation data from SMRT-BS was validated against overlapping CpG site methylation data derived from the Infinium HumanMethylation450 BeadChip (450K-array; Illumina, Inc., San Diego, CA, USA). The samples were processed according to the manufacturer’s instructions by the Genomics Core Laboratory at the Icahn School of Medicine at Mount Sinai without any modification to the manufacturer’s protocol. 450K-array data were obtained and normalized using GenomeStudio® software (Illumina) according to the manufacturer’s instructions and the Illumina Methylation Analyzer (IMA) R package . 450K-array probes with detection p-values greater than 0.01 or with missing values were removed prior to further analyses. Multiple-mapped probes with less than two mismatches or indels were identified by the Bowtie aligner and removed to avoid potential errors of unspecific binding. In addition, probes that included known single-nucleotide polymorphisms according to 450K-array annotation were removed and peak-based normalization was performed using IMA.
SureSelect™ human Methyl-Seq
Methylation data from SMRT-BS was also validated against overlapping CpG site methylation data derived from the SureSelect™ Human Methyl-Seq (Agilent Technologies, Santa Clara, CA, USA) target enrichment system. Libraries were prepared according to the manufacturer’s instructions without any modification to the manufacturer’s protocol. In brief, 2 μg of genomic DNAs were sheared by a Covaris E210 to fragments of ~150 bp, end-repaired, 3’ end-adenylated and ligated with methylated adaptors. Pre-prepared DNA fragments were hybridized to SureSelect™ Human Methyl-Seq capture libraries, and target captured DNAs were treated with bisulfite, subjected to eight PCR cycles to enrich adaptor added fragments and six PCR cycles to add multiplexing barcodes. Sequencing was accomplished using a HiSeq 2000 (Illumina) and 100 bp single-read sequencing with one lane per two samples. Raw sequencing data was analyzed with Bismark  to extract DNA methylation information.
Availability of supporting data
Data sets supporting the results of this study were uploaded to public data repositories for open access. The SMRT-BS sequencing FASTQ data are available from the NCBI Sequence Read Archive (SRA) using the SRX977650 and SRX977540 experiment identifiers: http://www.ncbi.nlm.nih.gov/sra/SRX977650[accn] and http://www.ncbi.nlm.nih.gov/sra/SRX977540[accn] . The 450K-array and Methyl-Seq validation data are available from the LabArchives Electronic Laboratory Notebook (DOI: 10.6070/H45H7D8C).
Circular consensus sequencing
Reduced representation bisulfite sequencing
SMRT bisulfite sequencing
The authors would like to thank Karen Mochoruk for technical laboratory assistance. This work was supported in part through the computational resources and staff expertise provided by the Department of Scientific Computing at the Icahn School of Medicine at Mount Sinai.
- Jaenisch R, Bird A. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet. 2003;33(Suppl):245–54.View ArticlePubMedGoogle Scholar
- Mohandas T, Sparkes RS, Shapiro LJ. Reactivation of an inactive human X chromosome: evidence for X inactivation by DNA methylation. Science. 1981;211:393–6.View ArticlePubMedGoogle Scholar
- Baylin SB, Herman JG. DNA hypermethylation in tumorigenesis: epigenetics joins genetics. Trends Genet. 2000;16:168–74.View ArticlePubMedGoogle Scholar
- Li E, Beard C, Jaenisch R. Role for DNA methylation in genomic imprinting. Nature. 1993;366:362–5.View ArticlePubMedGoogle Scholar
- Meissner A, Gnirke A, Bell GW, Ramsahoye B, Lander ES, Jaenisch R. Reduced representation bisulfite sequencing for comparative high-resolution DNA methylation analysis. Nucleic Acids Res. 2005;33:5868–77.View ArticlePubMed CentralPubMedGoogle Scholar
- Yang Y, Wang W, Li Y, Tu J, Bai Y, Xiao P, et al. Identification of methylated regions with peak search based on Poisson model from massively parallel methylated DNA immunoprecipitation-sequencing data. Electrophoresis. 2010;31:3537–44.View ArticlePubMedGoogle Scholar
- Fouse SD, Nagarajan RO, Costello JF. Genome-scale DNA methylation analysis. Epigenomics. 2010;2:105–17.View ArticlePubMed CentralPubMedGoogle Scholar
- Plongthongkum N, Diep DH, Zhang K. Advances in the profiling of DNA modifications: cytosine methylation and beyond. Nat Rev Genet. 2014;15:647–61.View ArticlePubMedGoogle Scholar
- Shanmuganathan R, Basheer NB, Amirthalingam L, Muthukumar H, Kaliaperumal R, Shanmugam K. Conventional and nanotechniques for DNA methylation profiling. J Mol Diagn. 2013;15:17–26.View ArticlePubMedGoogle Scholar
- Frommer M, McDonald LE, Millar DS, Collis CM, Watt F, Grigg GW, et al. A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proc Natl Acad Sci U S A. 1992;89:1827–31.View ArticlePubMed CentralPubMedGoogle Scholar
- Choufani S, Shapiro JS, Susiarjo M, Butcher DT, Grafodatskaya D, Lou Y, et al. A novel approach identifies new differentially methylated regions (DMRs) associated with imprinted genes. Genome Res. 2011;21:465–76.View ArticlePubMed CentralPubMedGoogle Scholar
- Bliek J, Verde G, Callaway J, Maas SM, De Crescenzo A, Sparago A, et al. Hypomethylation at multiple maternally methylated imprinted regions including PLAGL1 and GNAS loci in Beckwith-Wiedemann syndrome. Eur J Hum Genet. 2009;17:611–9.View ArticlePubMed CentralPubMedGoogle Scholar
- Uhlmann K, Brinckmann A, Toliat MR, Ritter H, Nurnberg P. Evaluation of a potential epigenetic biomarker by quantitative methyl-single nucleotide polymorphism analysis. Electrophoresis. 2002;23:4072–9.View ArticlePubMedGoogle Scholar
- Tost J, El abdalaoui H, Gut IG. Serial pyrosequencing for quantitative DNA methylation analysis. Biotechniques. 2006;40:721–2. 724, 726.View ArticlePubMedGoogle Scholar
- Taylor KH, Kramer RS, Davis JW, Guo J, Duff DJ, Xu D, et al. Ultradeep bisulfite sequencing analysis of DNA methylation patterns in multiple gene promoters by 454 sequencing. Cancer Res. 2007;67:8511–8.View ArticlePubMedGoogle Scholar
- Krueger F, Kreck B, Franke A, Andrews SR. DNA methylome analysis using short bisulfite sequencing data. Nat Methods. 2012;9:145–51.View ArticlePubMedGoogle Scholar
- Paul DS, Guilhamon P, Karpathakis A, Butcher LM, Thirlwell C, Feber A, et al. Assessment of RainDrop BS-seq as a method for large-scale, targeted bisulfite sequencing. Epigenetics. 2014;9:678–84.View ArticlePubMed CentralPubMedGoogle Scholar
- Tost J, Gut IG. DNA methylation analysis by pyrosequencing. Nat Protoc. 2007;2:2265–75.View ArticlePubMedGoogle Scholar
- Booth MJ, Ost TW, Beraldi D, Bell NM, Branco MR, Reik W, et al. Oxidative bisulfite sequencing of 5-methylcytosine and 5-hydroxymethylcytosine. Nat Protoc. 2013;8:1841–51.View ArticlePubMed CentralPubMedGoogle Scholar
- Stewart SK, Morris TJ, Guilhamon P, Bulstrode H, Bachman M, Balasubramanian S, et al. oxBS-450 K: A method for analysing hydroxymethylation using 450 K BeadChips. Methods. 2015;72:9–15.View ArticlePubMed CentralPubMedGoogle Scholar
- Gu H, Smith ZD, Bock C, Boyle P, Gnirke A, Meissner A. Preparation of reduced representation bisulfite sequencing libraries for genome-scale DNA methylation profiling. Nat Protoc. 2011;6:468–81.View ArticlePubMedGoogle Scholar
- Adusumalli S, Mohd Omar MF, Soong R, Benoukraf T: Methodological aspects of whole-genome bisulfite sequencing analysis. Brief Bioinform 2014. [Epub ahead of print]
- Eid J, Fehr A, Gray J, Luong K, Lyle J, Otto G, et al. Real-time DNA sequencing from single polymerase molecules. Science. 2009;323:133–8.View ArticlePubMedGoogle Scholar
- Grunau C, Clark SJ, Rosenthal A. Bisulfite genomic sequencing: systematic investigation of critical experimental parameters. Nucleic Acids Res. 2001;29:E65–65.View ArticlePubMed CentralPubMedGoogle Scholar
- Warnecke PM, Stirzaker C, Song J, Grunau C, Melki JR, Clark SJ. Identification and resolution of artifacts in bisulfite sequencing. Methods. 2002;27:101–7.View ArticlePubMedGoogle Scholar
- Tusnady GE, Simon I, Varadi A, Aranyi T. BiSearch: primer-design and search tool for PCR on bisulfite-treated genomes. Nucleic Acids Res. 2005;33, e9.View ArticlePubMed CentralPubMedGoogle Scholar
- Li Y, Tollefsbol TO. DNA methylation detection: bisulfite genomic sequencing analysis. Methods Mol Biol. 2011;791:11–21.View ArticlePubMed CentralPubMedGoogle Scholar
- Su XZ, Wu Y, Sifri CD, Wellems TE. Reduced extension temperatures required for PCR amplification of extremely A + T-rich DNA. Nucleic Acids Res. 1996;24:1574–5.View ArticlePubMed CentralPubMedGoogle Scholar
- Rohde C, Zhang Y, Reinhardt R, Jeltsch A. BISMA–fast and accurate bisulfite sequencing data analysis of individual clones from unique and repetitive sequences. BMC Bioinformatics. 2010;11:230.View ArticlePubMed CentralPubMedGoogle Scholar
- Warnecke PM, Stirzaker C, Melki JR, Millar DS, Paul CL, Clark SJ. Detection and measurement of PCR bias in quantitative methylation analysis of bisulphite-treated DNA. Nucleic Acids Res. 1997;25:4422–6.View ArticlePubMed CentralPubMedGoogle Scholar
- Varley KE, Gertz J, Bowling KM, Parker SL, Reddy TE, Pauli-Behn F, et al. Dynamic DNA methylation across diverse human cell lines and tissues. Genome Res. 2013;23:555–67.View ArticlePubMed CentralPubMedGoogle Scholar
- Flusberg BA, Webster DR, Lee JH, Travers KJ, Olivares EC, Clark TA, et al. Direct detection of DNA methylation during single-molecule, real-time sequencing. Nat Methods. 2010;7:461–5.View ArticlePubMed CentralPubMedGoogle Scholar
- Moskalev EA, Zavgorodnij MG, Majorova SP, Vorobjev IA, Jandaghi P, Bure IV, et al. Correction of PCR-bias in quantitative DNA methylation studies by means of cubic polynomial regression. Nucleic Acids Res. 2011;39, e77.View ArticlePubMed CentralPubMedGoogle Scholar
- Michels KB, Binder AM, Dedeurwaerder S, Epstein CB, Greally JM, Gut I, et al. Recommendations for the design and analysis of epigenome-wide association studies. Nat Methods. 2013;10:949–55.View ArticlePubMedGoogle Scholar
- Paul DS, Beck S. Advances in epigenome-wide association studies for common diseases. Trends Mol Med. 2014;20:541–3.View ArticlePubMed CentralPubMedGoogle Scholar
- Gardiner-Garden M, Frommer M. CpG islands in vertebrate genomes. J Mol Biol. 1987;196:261–82.View ArticlePubMedGoogle Scholar
- Scott SA, Khasawneh R, Peter I, Kornreich R, Desnick RJ. Combined CYP2C9, VKORC1 and CYP4F2 frequencies among racial and ethnic groups. Pharmacogenomics. 2010;11:781–91.View ArticlePubMed CentralPubMedGoogle Scholar
- Rasko DA, Webster DR, Sahl JW, Bashir A, Boisen N, Scheutz F, et al. Origins of the E. coli strain causing an outbreak of hemolytic-uremic syndrome in Germany. N Engl J Med. 2011;365:709–17.View ArticlePubMed CentralPubMedGoogle Scholar
- Breese MR, Liu Y. NGSUtils: a software suite for analyzing and manipulating next-generation sequencing datasets. Bioinformatics. 2013;29:494–6.View ArticlePubMed CentralPubMedGoogle Scholar
- Krueger F, Andrews SR. Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics. 2011;27:1571–2.View ArticlePubMed CentralPubMedGoogle Scholar
- Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012;9:357–9.View ArticlePubMed CentralPubMedGoogle Scholar
- Wang D, Yan L, Hu Q, Sucheston LE, Higgins MJ, Ambrosone CB, et al. IMA: an R package for high-throughput analysis of Illumina’s 450 K Infinium methylation data. Bioinformatics. 2012;28:729–30.View ArticlePubMed CentralPubMedGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.