Dataset Information

A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450 k DNA methylation data.

ABSTRACT: The Illumina Infinium 450 k DNA Methylation Beadchip is a prime candidate technology for Epigenome-Wide Association Studies (EWAS). However, a difficulty associated with these beadarrays is that probes come in two different designs, characterized by widely different DNA methylation distributions and dynamic range, which may bias downstream analyses. A key statistical issue is therefore how best to adjust for the two different probe designs.Here we propose a novel model-based intra-array normalization strategy for 450 k data, called BMIQ (Beta MIxture Quantile dilation), to adjust the beta-values of type2 design probes into a statistical distribution characteristic of type1 probes. The strategy involves application of a three-state beta-mixture model to assign probes to methylation states, subsequent transformation of probabilities into quantiles and finally a methylation-dependent dilation transformation to preserve the monotonicity and continuity of the data. We validate our method on cell-line data, fresh frozen and paraffin-embedded tumour tissue samples and demonstrate that BMIQ compares favourably with two competing methods. Specifically, we show that BMIQ improves the robustness of the normalization procedure, reduces the technical variation and bias of type2 probe values and successfully eliminates the type1 enrichment bias caused by the lower dynamic range of type2 probes. BMIQ will be useful as a preprocessing step for any study using the Illumina Infinium 450 k platform.BMIQ is freely available from http://code.google.com/p/bmiq/.a.teschendorff@ucl.ac.ukSupplementary data are available at Bioinformatics online.

SUBMITTER: Teschendorff AE

PROVIDER: S-EPMC3546795 | biostudies-literature | 2013 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450 k DNA methylation data.

Teschendorff Andrew E AE Marabita Francesco F Lechner Matthias M Bartlett Thomas T Tegner Jesper J Gomez-Cabrero David D Beck Stephan S

Bioinformatics (Oxford, England) 20121121 2

<h4>Motivation</h4>The Illumina Infinium 450 k DNA Methylation Beadchip is a prime candidate technology for Epigenome-Wide Association Studies (EWAS). However, a difficulty associated with these beadarrays is that probes come in two different designs, characterized by widely different DNA methylation distributions and dynamic range, which may bias downstream analyses. A key statistical issue is therefore how best to adjust for the two different probe designs.<h4>Results</h4>Here we propose a nov ...[more]

PMID: 23175756

Similar Datasets

Project description:Hepatocellular carcinoma (HCC) incidence has increased in the US and also has one of the fastest growing death rates of any cancer. The purpose of the current study was to discover novel genome-wide aberrant DNA methylation patterns in HCC tumors that are predominantly HCV-related. Infinium HumanMethylation 450K BeadChip arrays were used to examine genome-wide DNA methylation profiles in 66 pairs of HCC tumor and adjacent non-tumor tissues. After Bonferroni adjustment, a total of 130,512 CpG sites significantly differed in methylation level in tumor compared with non-tumor tissues, with 28,017 CpG sites hypermethylated and 102,495 hypomethylated in tumor tissues. Absolute tumor/non-tumor methylation differences ? 20% were found in 24.9% of the hypermethylated and 43.1% of the hypomethylated CpG sites; almost 10,000 CpG sites have ? 30% DNA methylation differences. Most (60.1%) significantly hypermethylated CpG sites are located in CpG islands, with 21.6% in CpG shores and 3.6% in shelves. In contrast, only a small proportion (8.2%) of significantly hypomethylated CpG sites are situated in islands, while most are found in open sea (60.2%), shore (17.3%) or shelf (14.3%) regions. A total of 2,568 significant CpG sites (2,441 hypermethylated and 127 hypomethylated) covering 589 genes are located within 684 differentially methylated regions defined as regions with at least two significant CpG sites displaying > 20% methylation differences in the same direction within 250-bp. The top 500 significant CpG sites can significantly distinguish HCC tumor from adjacent tissues with one misclassification. Within adjacent non-tumor tissues, we also identified 75 CpG sites significantly associated with gender, 228 with HCV infection, 17,207 with cirrhosis, and 56 with both HCV infection and cirrhosis after multiple comparisons adjustment. Aberrant DNA methylation profiles across the genome were identified in tumor tissues from US HCC cases that are predominantly related to HCV infection. These results demonstrate the significance of aberrant DNA methylation in HCC tumorigenesis.

Project description:BACKGROUND: Genome-wide methylation profiling has led to more comprehensive insights into gene regulation mechanisms and potential therapeutic targets. Illumina Human Methylation BeadChip is one of the most commonly used genome-wide methylation platforms. Similar to other microarray experiments, methylation data is susceptible to various technical artifacts, particularly batch effects. To date, little attention has been given to issues related to normalization and batch effect correction for this kind of data. METHODS: We evaluated three common normalization approaches and investigated their performance in batch effect removal using three datasets with different degrees of batch effects generated from HumanMethylation27 platform: quantile normalization at average ? value (QN?); two step quantile normalization at probe signals implemented in "lumi" package of R (lumi); and quantile normalization of A and B signal separately (ABnorm). Subsequent Empirical Bayes (EB) batch adjustment was also evaluated. RESULTS: Each normalization could remove a portion of batch effects and their effectiveness differed depending on the severity of batch effects in a dataset. For the dataset with minor batch effects (Dataset 1), normalization alone appeared adequate and "lumi" showed the best performance. However, all methods left substantial batch effects intact in the datasets with obvious batch effects and further correction was necessary. Without any correction, 50 and 66 percent of CpGs were associated with batch effects in Dataset 2 and 3, respectively. After QN?, lumi or ABnorm, the number of CpGs associated with batch effects were reduced to 24, 32, and 26 percent for Dataset 2; and 37, 46, and 35 percent for Dataset 3, respectively. Additional EB correction effectively removed such remaining non-biological effects. More importantly, the two-step procedure almost tripled the numbers of CpGs associated with the outcome of interest for the two datasets. CONCLUSION: Genome-wide methylation data from Infinium Methylation BeadChip can be susceptible to batch effects with profound impacts on downstream analyses and conclusions. Normalization can reduce part but not all batch effects. EB correction along with normalization is recommended for effective batch effect removal.

Project description:microRNA (miRNA) expression profiles based on the highly powerful Illumina sequencing technology rely on the construction of cDNA libraries in which adaptor ligation is known to deeply favor some miRNAs over others. This introduces erroneous measurements of the miRNA abundances and relative miRNA quantities in biological samples. Here, by using the commercial miRXplore Universal Reference that contains an equimolar mixture of 963 animal miRNAs and TruSeq or bulged adaptors, we describe a method for correcting ligation biases in expression profiles obtained with standard protocols of cDNA library construction and provide data for quantifying the true miRNA abundances in biological samples. Ligation biases were evaluated at three ratios of miRNA to 3'-adaptor and four numbers of polymerase chain reaction amplification cycles by calculating efficiency captures/correcting factors for each miRNA. We show that ligation biases lead to over- or under-expression covering a 105 amplitude range. We also show that, at each miRNA:3'-adaptor ratio, coefficients of variation (CVs) of efficiency captures calculated over the four number of amplification cycles using sliding windows of 10 values ranged from 0.1 for the miRNAs of high expression to 0.6 for the miRNAs of low expression. Efficiency captures of miRNAs of high and low expression in profiles are therefore differently impacted by the number of amplification cycles. Importantly, we observed that at a given number of amplification cycles, CVs of efficiency captures calculated over the three miRNA:3'-adaptor ratios displayed a steady value of 0.3 +/- 0.05 STD for miRNAs of high and low expression. This allows, at a given number of amplification cycles, accurate comparison of miRNA expression between biological samples over a substantial expression range. Finally we provide tables of correcting factors that allow to measure the abundances of 963 miRNAs in biological samples from TruSeq-based expression profiles and, an example of their use by characterizing miRNAs of the let-7, miR-26, miR-29, and miR-30 families as the more abundant miRNAs of the rat adult cerebellum.

Dataset Information

A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450 k DNA methylation data.

Publications

A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450 k DNA methylation data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets