Interpretation of genome-wide infinium methylation data from ligated DNA in formalin-fixed, paraffin-embedded paired tumor and normal tissue
© Jasmine et al; licensee BioMed Central Ltd. 2012
Received: 7 November 2011
Accepted: 22 February 2012
Published: 22 February 2012
Formalin-fixed, paraffin-embedded (FFPE) samples are a highly desirable resource for epigenetic studies, but there is no suitable platform to assay genome-wide methylation in these widely available resources. Recently, Thirlwell et al. (2010) have reported a modified ligation-based DNA repair protocol to prepare FFPE DNA for the Infinium methylation assay. In this study, we have tested the accuracy of methylation data obtained with this modification by comparing paired fresh-frozen (FF) and FFPE colon tissue (normal and tumor) from colorectal cancer patients. We report locus-specific correlation and concordance of tumor-specific differentially methylated loci (DML), both of which were not previously assessed.
We used Illumina's Infinium Methylation 27K chip for 12 pairs of FF and 12 pairs of FFPE tissue from tumor and surrounding healthy tissue from the resected colon of the same individual, after repairing the FFPE DNA using Thirlwell's modified protocol.
For both tumor and normal tissue, overall correlation of β values between all loci in paired FF and FFPE was comparable to previous studies. Tissue storage type (FF or FFPE) was found to be the most significant source of variation rather than tissue type (normal or tumor). We found a large number of DML between FF and FFPE DNA. Using ANOVA, we also identified DML in tumor compared to normal tissue in both FF and FFPE samples, and out of the top 50 loci in both groups only 7 were common, indicating poor concordance. Likewise, while looking at the correlation of individual loci between FFPE and FF across the patients, less than 10% of loci showed strong correlation (r ≥ 0.6). Finally, we checked the effect of the ligation-based modification on the Infinium chemistry for SNP genotyping on an independent set of samples, which also showed poor performance.
Ligation of FFPE DNA prior to the Infinium genome-wide methylation assay may detect a reasonable number of loci, but the numbers of detected loci are much fewer than in FF samples. More importantly, the concordance of DML detected between FF and FFPE DNA is suboptimal, and DML from FFPE tissues should be interpreted with great caution.
Aberrant DNA methylation is a well-established pathway in carcinogenesis [1, 2]. In colorectal cancer (CRC), global hypomethylation of DNA and gene-specific hypermethylation of tumor suppressor genes and microRNA genes are extensively studied . For example, seminal work in cancer epigenetics has shown that most cases of microsatellite-instable CRC are caused by the hypermethylation and consequent silencing of the mismatch-repair gene MLH1 [4, 5]. Many epigenetic markers for CRC are now known, including MGMT [6, 7], VIM , APC , RUNX3 [5, 10], CDKN2A , and numerous others found in recent genome-wide studies [12–14]. It is hoped that continuing studies can provide useful strategies for detection, treatment, and the understanding of etiology .
Formalin-fixed, paraffin-embedded (FFPE) samples are routinely collected for histopathological diagnosis and are thus a highly desirable resource for epigenetic studies. Though formalin fixation does not alter the methylation status of cytosine , it does cause other forms of DNA damage, including cross-linking, fragmentation, and generation of apurinic/apyrimidinic sites . This degradation can be detrimental to qPCR  or whole-genome amplification (WGA) , which are integral steps in many methylation assays.
Therefore, any existing methylation assay must be carefully evaluated before it can be confidently used for FFPE-derived DNA. Many methylation assays have been evaluated for such purposes [20–26]. The most comprehensive validations involve comparisons between paired FFPE and fresh-frozen (FF) tissue samples, such as the validations reported for: high-resolution melting analysis , qPCR quantification after methylation-specific restriction enzyme digestion , bisulfite sequencing , and Illumina's GoldenGate methylation assay . Killian's validation of the GoldenGate assay showed good correlation between paired FFPE and FF samples but the GoldenGate assay interrogates a limited number of CpG loci and is not suitable for studies of large numbers of loci . For genome-wide studies, Illumina's Infinium assay  allows thousands of loci to be interrogated at a time. However, the Infinium chemistry depends on WGA and thus was originally designed for high-quality, high molecular weight DNA. Many WGA protocols fail with fragmented FFPE DNA , and Illumina's proprietary Infinium WGA chemistry has been shown to fail with these samples by both in-house  and independent results . GoldenGate chemistry, on the other hand, does not involve WGA, targets small fragments of DNA for PCR amplification, and thus may be compatible with FFPE DNA. In a previous study we compared genome-wide gene expression (using Illumina's DASL assay, based on GoldenGate chemistry) in paired FF and FFPE breast tumor tissues and surrounding healthy tissues . In that study, we found that the tumor specific differentially expressed genes detected in FF and FFPE samples were significantly different, suggesting that interpreting FFPE gene expression data may be problematic.
Recently, Thirlwell et al. described a modified Infinium methylation protocol in which FFPE DNA was repaired by ligation prior to the bisulfite conversion and methylation assay . Thirlwell's protocol was shown to be effective in several respects. First, the authors showed that ligation allowed successful WGA, whereas unligated replicates failed to amplify. Second, the authors demonstrated reasonable correlation between paired FF and FFPE samples from primary ovarian cancer tissues. However, they did not report whether FFPE tumor samples could detect the same differentially methylated loci (DML) as were detected by examining FF tumor tissue. Recently, we reported results from a genome-wide DNA methylation study in colorectal cancer using Illumina's Infinium-based HumanMethylation27 microarray . The study was conducted using paired FF tumor and adjacent normal colon tissue samples from 24 patients. FFPE tumor and normal samples were also available from the same patients. This provided an excellent opportunity to independently test the utility of ligated-FFPE DNA for Infinium methylation analysis using Thirlwell's modification . In the current study, we have tested the accuracy of methylation data from ligated-FFPE DNA through numerous correlations with paired FF DNA for both CRC and adjacent normal colon tissue. We also identified DML in FF tumor DNA (compared to FF adjacent normal) and compared these to loci that were differentially methylated in ligated-FFPE tumor DNA (compared to ligated-FFPE adjacent normal) from the same patients. Our study is unique amongst other validations of FFPE methylation assays in that we have reported results for locus-specific correlations across samples and concordance of tumor-specific DML.
Colon tissues (tumor and surrounding healthy) were collected from surgically removed colonic segments from consecutive patients at Bangabandhu Sheikh Mujib Medical University (BSMMU), Dhaka, Bangladesh, as described previously . All samples were collected by one surgical pathology fellow (MR) from the operating room immediately after surgical resection during the period of December 2009 to March 2010. Histopathology was done independently by two histopathologists (MK & MR), and there was concordance in all cases. For each patient, one sample was collected from the tumor mass, and another sample was taken from the resected, unaffected part of the colon about 5-10 cm away from the tumor mass. From each site, tissue sections were preserved as (1) fresh frozen, (2) in RNA-stabilizing buffer and (3) as FFPE block. The samples were shipped on dry ice to the molecular genomics lab at the University of Chicago for subsequent DNA extraction and methylation assay. We also received the corresponding FFPE blocks that were used for histopathology. Written informed consent was obtained from all participants. The research protocol was approved by the "Ethical Review Committee, Bangabandhu Sheikh Mujib Medical University", Dhaka, Bangladesh (BSMMU/2010/10096) and by the "Biological Sciences Division, University of Chicago Hospital Institutional Review Board", Chicago, IL, USA (10-264-E). We have previously reported genome-wide methylation data from the first 24 paired (tumor and corresponding healthy colonic tissue) FF DNA . In this paper we present methylation data from FFPE sections of the first 12 consecutive patients of the same series for whom we had paired (normal and CRC) FFPE blocks available and compared the data with corresponding 12 pairs from DNA from FF samples.
DNA extraction and quality control
DNA was extracted from FFPE tissue (tumor and surrounding healthy tissue) using the Puregene Core kit A (Qiagen, Maryland, USA). During extraction all DNA samples were treated with RNase. FFPE tissues were about 1 year old. All DNA concentrations were measured by Nanodrop (Thermo-Fisher, USA), and integrity was checked by the Agilent Bioanalyzer 2100 using the DNA 12000 kit (Agilent Technologies, USA).
2 μg FFPE DNA were ligated before starting bisulfite conversion using the protocol described by Thirlwell et al. . For bisulfite conversion, the EZ DNA methylation kit (Zymo Research, USA) was used.
Genome-wide methylation assay
The Infinium Methylation assay (Illumina Inc., USA) was done using the Methylation 27K chip, which contains 27,578 CpG sites spanning 14,495 genes. The CpG sites were located within the proximal promoter regions of genes, with the distance to transcription start site (TSS) ranging from 0 to 1499 bp and averaged at 389 ± 341 bp. Paired FFPE DNA from CRC and surrounding normal colonic tissues were processed on the same chip to avoid batch effects, and all 24 FFPE samples were processed on 2 chips (12 samples per chip). It may be noted that the corresponding 24 FF samples were processed in a different batch previously, but the corresponding DNA samples from normal and CRC tissue were processed in the same chip. A Tecan Evo robot was used for automated sample processing and the chips were scanned on a single BeadArray reader (S428). Illumina's BeadStudio analytical software showed excellent intensity for staining (above 15000), clear clustering for the hybridization probes, good target removal intensity (< 400) and satisfactory bisulfite conversion.
Genome-wide methylation data analysis
Where Yijklmn represents the nth observation on the ith Tissue, jth Age_Cat40, kth Sex, lth Location and mth Person; μ is the common effect for the whole experiment, εijklmn represents the random error present in the nth observation on the ith Tissue, jth Age_Cat40, kth Sex, lth Location, and mth Person. The errors εijklmn are assumed to be normally and independently distributed with mean 0 and standard deviation δ for all measurements. An FDR of 0.05 was used for multiple testing correction.
The correlation of β values between FF and FFPE samples from each individual was checked in both normal and tumor tissue. Then the correlation of the average β values of all the FF and corresponding FFPE samples was analyzed both for normal and tumor tissue.
The distribution of the genes correlated between FF and FFPE samples were also checked. The top 50 differentially methylated genes between normal and tumor tissue in both FF and FFPE tissue were detected to find the common genes.
Detection of loci
Sources of variation in methylation
Histograms of methylation β-values by storage type
Correlation between storage-type and tissue-type pairs
Then the β values of all 27,578 loci from each tumor sample were plotted against the β values from the corresponding normal tissue of the same patient. Representative scatter plots from one patient (C_1) are shown in Figure 4C (for FF) and 4D (for FFPE). As expected, the scatter plots indicate differential methylation of a number of loci in CRC tissue compared to normal tissue.
Similarly, we looked at the correlation of mean β values of tumor FF tissue vs. mean β-values of normal FF tissue (r2 = 0.93, Figure 5C); and mean β values of tumor FFPE tissue vs. mean β values of normal FFPE tissue (r2 = 0.91, Figure 5D). Average tumor vs. normal β values showed better correlation than individual sample pairs, for both FF and FFPE tissue. This increase in r2 is related to the direct increase in data points used for analysis.
Locus-specific correlation between FF and FFPE DNA
Differentially methylated loci (DML) in FFPE compared to FF
Differentially methylated loci in tumor tissue
Our data suggests that data generated by Thirlwell's modified Infinium methylation protocol can identify > 95% loci in FFPE samples, which is significantly fewer than what is seen in FF samples in the unmodified Infinium protocol. Based on these data, it is also possible to separate the tumor and normal samples. In fact, DML sets from FF and FFPE tissue are both effective at differentiating tumor and normal tissue (Figure 8A and 8B). However, these DML sets are very discordant, suggesting that using the modified Infinium assay with FFPE samples may not provide the same biological information as the unmodified assay with FF samples (the gold standard in this case). Furthermore, the DML from FFPE show a greater amount of variation that cannot be attributed to differences in tissue type (Figure 9).
Our study has attempted to validate Thirlwell's modified Infinium protocol by using 12 pairs FF and FFPE samples from 12 primary CRC samples and 12 adjacent normal tissues. Thirlwell et al.  compared 2 FF ovarian cancer tissue with paired FFPE DNA from ligated and unligated replicates. The authors showed good correlation of β values and intensity between FF and ligated FFPE DNA compared to unligated FFPE DNA. However, Thirlwell did not report changes in differential methylation that are typically investigated in cancer research - for example, whether ligated-FFPE DNA produced the same DML sets as FF DNA, or whether ligated-FFPE DNA had the same power to distinguish tumor samples from normal samples. This kind of hypothesis testing is of particular concern since we have previously reported that in Illumina's DASL whole-genome gene expression microarray, FFPE RNA can yield significantly different results compared to paired FF RNA .
In our study, the overall correlation of β values were comparable to Thirlwell's study . When looking at the correlation between FF and FFPE biological replicates, Thirlwell found a median r2 = 0.91, range = 0.88 - 0.96. This was slightly higher than our observed correlations, but not as high as correlations reported by Killian for the GoldenGate assay , which ranged from r2 = 0.95 to 0.99, and it may be noted that GoldenGate chemistry is suitable for FFPE samples whereas in principle Infinium chemistry is not. However, our data suggests that correlation is definitely related to the number of data points in the analysis.
Ideally, our experiment would have been designed to include all the four samples from same patient (tumor and normal from FF and FFPE sections) on the same chip to totally eliminate any batch effects. Unfortunately, the FF samples were processed earlier for a separate study. However, in both the cases (FF and FFPE) paired tumor-normal samples were processed on the same chip. Since the β value is calculated from the ratio of the signal intensity values (methylated to total), slight differences in intensities are less likely to affect β.
A few studies have evaluated the use of FFPE tissue was evaluated for lower throughput methylation assays with fewer CpG locations. Balic et al.  used HRM to interrogate promoter methylation of two genes (MGMT and APC), and compared results from paired FFPE and FF samples in 5 human breast cancer cell lines and 3 human prostate cancer cell lines; these results were also validated with the MethyLight qPCR assay. Gagnon et al. validated promoter methylation status of PLAU and TIMP3 genes in FFPE tissue using methylation sensitive restriction enzyme digestion and qPCR; this was done for paired FFPE and FF samples from 9 primary breast tumor samples and 4 cell line admixtures . Killian et al. evaluated the GoldenGate methylation assay on paired FF and FFPE tissue from 10 lymphoma samples and 10 lymph node hyperplasia samples . They found good correlation of DML between FF and FFPE in different groups, although the number of loci was small. Even though Killian identified lymphoma-specific DML in comparison with hyperplasia samples, the lymphoma and hyperplasia samples were not paired from the same patient (unlike our study, in which tumor and adjacent normal samples were paired).
To our knowledge, our study is unique in addressing this issue at a genome-wide level using a large number of samples and a well-designed experiment to validate tumor-specific DML data derived from paired FFPE DNA (tumor and normal) against DML data from paired corresponding FF tissue DNA. The discrepancy between FFPE and FF samples in our study may reflect: (a) the incompatibility of the Infinium chemistry, including the WGA component; and/or (b) DNA damage induced by tissue fixation, which may lead to misidentification or miscalculation of β values. Fixation-induced changes in methylation status may be ruled out, since previous authors' GoldenGate data and other low-plex methylation data did not find significant differences between FFPE samples and corresponding FF samples.
In conclusion, ligation-based repair of FFPE DNA may allow the Infinium whole-genome methylation assay to detect a reasonable number of loci, although much fewer than in unmodified FF DNA. Infinium methylation data from ligated FFPE DNA may also differentiate tumor and normal samples, but the DML sets derived from FFPE and FF samples are very discordant and may not provide the same biological information. Therefore, tumor-specific DML identified in FFPE tissue with this method should be interpreted with great caution.
This work was supported by the National Institutes of Health grants U01 CA122171, P30 CA 014599, P42ES010349, R01CA102484, and R01CA107431.
- Esteller M: Epigenetics in cancer. N Eng J Med. 2008, 358: 1148-1159. 10.1056/NEJMra072067.View ArticleGoogle Scholar
- Jones PA, Baylin SB: The fundamental role of epigenetic events in cancer. Nat Rev Genet. 2002, 3: 415-428.PubMedView ArticleGoogle Scholar
- Kim MS, Lee J, Sidransky D: DNA methylation markers in colorectal cancer. Cancer Metastasis Rev. 2010, 29: 181-206. 10.1007/s10555-010-9207-6.PubMedView ArticleGoogle Scholar
- Herman JG, Umar A, Polyak K, Graff JR, Ahuja N, Issa JPJ, Markowitz S, Willson JKV, Hamilton SR, Kinzler KW, et al: Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA. 1998, 95: 6870-6875. 10.1073/pnas.95.12.6870.PubMedPubMed CentralView ArticleGoogle Scholar
- Weisenberger DJ, DSiegmund K, Campan M, Young J, Long TI, Faasse MA, Kang GH, Widschwendter M, Weener D, Buchanan D, et al: CpG island methylator phenotype underlies sporadic microsatellite instability and is tightly associated with BRAF mutation in colorectal cancer. Nat Genet. 2006, 38: 787-793. 10.1038/ng1834.PubMedView ArticleGoogle Scholar
- Esteller M, Hamilton SR, Burger PC, Baylin SB, Herman JG: Inactivation of the DNA repair gene O-6-methylguanine-DNA methyltransferase by promoter hypermethylation is a common event in primary human neoplasia. Cancer Res. 1999, 59: 793-797.PubMedGoogle Scholar
- Esteller M, Risques RA, Toyota M, Capella G, Moreno V, Peinado MA, Baylin SB, Herman JG: Promoter hypermethylation of the DNA repair gene O-6-methylguanine-DNA methyltransferase is associated with the presence of G: C to A: T transition mutations in p53 in human colorectal tumorigenesis. Cancer Res. 2001, 61: 4689-4692.PubMedGoogle Scholar
- Chen WD, Han ZJ, Skoletsky J, Olson J, Sah J, Myeroff L, Platzer P, Lu SL, Dawson D, Willis J, et al: Detection in fecal DNA of colon cancer-specific methylation of the nonexpressed vimentin gene. J Natl Cancer Inst. 2005, 97: 1124-1132. 10.1093/jnci/dji204.PubMedView ArticleGoogle Scholar
- Esteller M, Sparks A, Toyota M, Sanchez-Cespedes M, Capella G, Peinado MA, Gonzalez S, Tarafa G, Sidransky D, Meltzer SJ, et al: Analysis of Adenomatous Polyposis Coli promoter hypermethylation in human cancer. Cancer Res. 2000, 60: 4366-4371.PubMedGoogle Scholar
- Ogino S, Kawasaki T, Kirkner GJ, Kraft P, Loda M, Fuchs CS: Evaluation of markers for CpG island methylator phenotype (CIMP) in colorectal cancer by a large population-based sample. J Mol Diagn. 2007, 9: 305-314. 10.2353/jmoldx.2007.060170.PubMedPubMed CentralView ArticleGoogle Scholar
- Herman JG, Baylin SB: Mechanisms of disease: gene silencing in cancer in association with promoter hypermethylation. N Eng J Med. 2003, 349: 2042-2054. 10.1056/NEJMra023075.View ArticleGoogle Scholar
- Kibriya MG, Raza M, Jasmine F, Roy S, Paul-Brutus R, Rahaman R, Dodsworth C, Rakibuz-Zaman M, Kamal M, Ahsan H: A genome-wide DNA methylation study in colorectal carcinoma. BMC Medical Genomics. 2011, 4: 50-10.1186/1755-8794-4-50.PubMedPubMed CentralView ArticleGoogle Scholar
- Kim Y-H, Lee HC, Kim S-Y, Il Yeom Y, Ryu KJ, Min B-H, Kim D-H, Son HJ, Rhee P-L, Kim JJ, et al: Epigenomic analysis of aberrantly methylated genes in colorectal cancer identifies genes commonly affected by epigenetic alterations. Ann Surg Oncol. 2011, 18: 2338-2347. 10.1245/s10434-011-1573-y.PubMedPubMed CentralView ArticleGoogle Scholar
- Øster B, Thorsen K, Lamy P, Wojdacz TK, Hansen LL, Birkenkamp-Demtröder K, Sørensen KD, Laurberg S, Ørntoft TF, Andersen CL: Identification and validation of highly frequent CpG island hypermethylation in colorectal adenomas and carcinomas. International Journal of Cancer. 2011, 129 (12): 2855-66. 10.1002/ijc.25951.View ArticleGoogle Scholar
- Feinberg AP: The epigenetics of cancer etiology. Semin Cancer Biol. 2004, 14: 427-432. 10.1016/j.semcancer.2004.06.005.PubMedView ArticleGoogle Scholar
- Kitazawa S, Kitazawa R, Maeda S: Identification of methylated cytosine from archival formalin-fixed paraffin-embedded specimens. Lab Invest. 2000, 80: 275-276. 10.1038/labinvest.3780031.PubMedView ArticleGoogle Scholar
- Gilbert MTP, Haselkorn T, Bunce M, Sanchez JJ, Lucas SB, Jewell LD, Van Marck E, Worobey M: The Isolation of nucleic acids from fixed, paraffin-embedded tissues-which methods are useful when?. Plos One. 2007, 2 (6): e537-10.1371/journal.pone.0000537.PubMedPubMed CentralView ArticleGoogle Scholar
- Lehmann U, Kreipe H: Real-time PCR analysis of DNA and RNA extracted from formalin-fixed and paraffin-embedded biopsies. Methods. 2001, 25: 409-418. 10.1006/meth.2001.1263.PubMedView ArticleGoogle Scholar
- Bosso M, Al-Mulla F: Whole genome amplification of DNA extracted from FFPE tissues formalin-fixed paraffin-embedded tissues. Methods Mol Biol. 2011, 724: 161-180. 10.1007/978-1-61779-055-3_11.PubMedView ArticleGoogle Scholar
- Gagnon J-F, Sanschagrin F, Jacob S, Tremblay A-A, Provencher L, Robert J, Morin C, Diorio C: Quantitative DNA methylation analysis of laser capture microdissected formalin-fixed and paraffin-embedded tissues. Exp Mol Pathol. 2010, 88: 184-189. 10.1016/j.yexmp.2009.09.020.PubMedView ArticleGoogle Scholar
- Gu H, Bock C, Mikkelsen TS, Jager N, Smith ZD, Tomazou E, Gnirke A, Lander ES, Meissner A: Genome-scale DNA methylation mapping of clinical samples at single-nucleotide resolution. Nat Methods. 2010, 7: 133-169. 10.1038/nmeth.1414.PubMedPubMed CentralView ArticleGoogle Scholar
- Irahara N, Nosho K, Baba Y, Shima K, Lindeman NI, Hazra A, Schernhammer ES, Hunter DJ, Fuchs CS, Ogino S: Precision of pyrosequencing assay to measure LINE-1 methylation in colon cancer, normal colonic mucosa, and peripheral blood cells. J Mol Diagn. 2010, 12: 177-183. 10.2353/jmoldx.2010.090106.PubMedPubMed CentralView ArticleGoogle Scholar
- Killian JK, Bilke S, Davis S, Walker RL, Killian MS, Jaeger EB, Chen Y, Hipp J, Pittaluga S, Raffeld M, et al: Large-scale profiling of archival lymph nodes reveals pervasive remodeling of the follicular lymphoma methylome. Cancer Res. 2009, 69: 758-764. 10.1158/0008-5472.CAN-08-2984.PubMedView ArticleGoogle Scholar
- Ogino S, Kawasaki T, Brahmandam M, Cantor M, Kirkner GJ, Spiegelman D, Makrigiorgos GM, Weisenberger DJ, Laird PW, Loda M, Fuchs CS: Precision and performance characteristics of bisulfite conversion and real-time PCR (MethyLight) for quantitative DNA methylation analysis. J Mol Diagn. 2006, 8: 209-217. 10.2353/jmoldx.2006.050135.PubMedPubMed CentralView ArticleGoogle Scholar
- Wojdacz TK, Dobrovic A: Methylation-sensitive high resolution melting (MS-HRM): a new approach for sensitive and high-throughput assessment of methylation. Nucleic Acids Res. 2007, 35: e41-10.1093/nar/gkm013.PubMedPubMed CentralView ArticleGoogle Scholar
- Wojdacz TK, Dobrovic A, Hansen LL: Methylation-sensitive high-resolution melting. Nat Protoc. 2008, 3: 1903-1908. 10.1038/nprot.2008.191.PubMedView ArticleGoogle Scholar
- Balic M, Pichler M, Strutz J, Heitzer E, Ausch C, Samonigg H, Cote RJ, Dandachi N: High quality assessment of DNA methylation in archival tissues from colorectal cancer patients using quantitative high-resolution melting analysis. J Mol Diagn. 2009, 11: 102-108. 10.2353/jmoldx.2009.080109.PubMedPubMed CentralView ArticleGoogle Scholar
- Bibikova M, Le J, Barnes B, Saedinia-Melnyk S, Zhou LX, Shen R, Gunderson KL: Genome-wide DNA methylation profiling using Infinium (R) assay. Epigenomics. 2009, 1: 177-200. 10.2217/epi.09.14.PubMedView ArticleGoogle Scholar
- Gunderson KL, Steemers FJ, Lee G, Mendoza LG, Chee MS: A genome-wide scalable SNP genotyping assay using microarray technology. Nat Genet. 2005, 37: 549-554. 10.1038/ng1547.PubMedView ArticleGoogle Scholar
- Thirlwell C, Eymard M, Feber A, Teschendorff A, Pearce K, Lechner M, Widschwendter M, Beck S: Genome-wide DNA methylation analysis of archival formalin-fixed paraffin-embedded tissue using the Illumina Infinium HumanMethylation27 BeadChip. Methods. 2010, 52: 248-254. 10.1016/j.ymeth.2010.04.012.PubMedView ArticleGoogle Scholar
- Kibriya M, Jasmine F, Roy S, Paul-Brutus R, Argos M, Ahsan H: Analyses and interpretation of whole-genome gene expression from formalin-fixed paraffin-embedded tissue: an illustration with breast cancer tissues. BMC Genomics. 2010, 11: 622-10.1186/1471-2164-11-622.PubMedPubMed CentralView ArticleGoogle Scholar
- Downey T: Analysis of a multifactor microarray study using partek genomics solution. Methods Enzymol. 2006, 411: 256-270.PubMedView ArticleGoogle Scholar
- Eisenhart C: The assumptions underlying the analysis of variance. Biometrics. 1947, 3: 1-21. 10.2307/3001534.PubMedView ArticleGoogle Scholar