Dataset Information

A comparison of feature selection and classification methods in DNA methylation studies using the Illumina Infinium platform.

ABSTRACT: The 27k Illumina Infinium Methylation Beadchip is a popular high-throughput technology that allows the methylation state of over 27,000 CpGs to be assayed. While feature selection and classification methods have been comprehensively explored in the context of gene expression data, relatively little is known as to how best to perform feature selection or classification in the context of Illumina Infinium methylation data. Given the rising importance of epigenomics in cancer and other complex genetic diseases, and in view of the upcoming epigenome wide association studies, it is critical to identify the statistical methods that offer improved inference in this novel context.Using a total of 7 large Illumina Infinium 27k Methylation data sets, encompassing over 1,000 samples from a wide range of tissues, we here provide an evaluation of popular feature selection, dimensional reduction and classification methods on DNA methylation data. Specifically, we evaluate the effects of variance filtering, supervised principal components (SPCA) and the choice of DNA methylation quantification measure on downstream statistical inference. We show that for relatively large sample sizes feature selection using test statistics is similar for M and ?-values, but that in the limit of small sample sizes, M-values allow more reliable identification of true positives. We also show that the effect of variance filtering on feature selection is study-specific and dependent on the phenotype of interest and tissue type profiled. Specifically, we find that variance filtering improves the detection of true positives in studies with large effect sizes, but that it may lead to worse performance in studies with smaller yet significant effect sizes. In contrast, supervised principal components improves the statistical power, especially in studies with small effect sizes. We also demonstrate that classification using the Elastic Net and Support Vector Machine (SVM) clearly outperforms competing methods like LASSO and SPCA. Finally, in unsupervised modelling of cancer diagnosis, we find that non-negative matrix factorisation (NMF) clearly outperforms principal components analysis.Our results highlight the importance of tailoring the feature selection and classification methodology to the sample size and biological context of the DNA methylation study. The Elastic Net emerges as a powerful classification algorithm for large-scale DNA methylation studies, while NMF does well in the unsupervised context. The insights presented here will be useful to any study embarking on large-scale DNA methylation profiling using Illumina Infinium beadarrays.

SUBMITTER: Zhuang J

PROVIDER: S-EPMC3364843 | biostudies-other | 2012 Apr

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

A comparison of feature selection and classification methods in DNA methylation studies using the Illumina Infinium platform.

Zhuang Joanna J Widschwendter Martin M Teschendorff Andrew E AE

BMC bioinformatics 20120424

<h4>Background</h4>The 27k Illumina Infinium Methylation Beadchip is a popular high-throughput technology that allows the methylation state of over 27,000 CpGs to be assayed. While feature selection and classification methods have been comprehensively explored in the context of gene expression data, relatively little is known as to how best to perform feature selection or classification in the context of Illumina Infinium methylation data. Given the rising importance of epigenomics in cancer and ...[more]

PMID: 22524302

Similar Datasets

Project description:BackgroundThere has been a steady increase in the number of studies aiming to identify DNA methylation differences associated with complex phenotypes. Many of the challenges of epigenetic epidemiology regarding study design and interpretation have been discussed in detail, however there are analytical concerns that are outstanding and require further exploration. In this study we seek to address three analytical issues. First, we quantify the multiple testing burden and propose a standard statistical significance threshold for identifying DNA methylation sites that are associated with an outcome. Second, we establish whether linear regression, the chosen statistical tool for the majority of studies, is appropriate and whether it is biased by the underlying distribution of DNA methylation data. Finally, we assess the sample size required for adequately powered DNA methylation association studies.ResultsWe quantified DNA methylation in the Understanding Society cohort (n = 1175), a large population based study, using the Illumina EPIC array to assess the statistical properties of DNA methylation association analyses. By simulating null DNA methylation studies, we generated the distribution of p-values expected by chance and calculated the 5% family-wise error for EPIC array studies to be 9 × 10- 8. Next, we tested whether the assumptions of linear regression are violated by DNA methylation data and found that the majority of sites do not satisfy the assumption of normal residuals. Nevertheless, we found no evidence that this bias influences analyses by increasing the likelihood of affected sites to be false positives. Finally, we performed power calculations for EPIC based DNA methylation studies, demonstrating that existing studies with data on ~ 1000 samples are adequately powered to detect small differences at the majority of sites.ConclusionWe propose that a significance threshold of P < 9 × 10- 8 adequately controls the false positive rate for EPIC array DNA methylation studies. Moreover, our results indicate that linear regression is a valid statistical methodology for DNA methylation studies, despite the fact that the data do not always satisfy the assumptions of this test. These findings have implications for epidemiological-based studies of DNA methylation and provide a framework for the interpretation of findings from current and future studies.

Project description:BackgroundThe field of epigenomics holds great promise in understanding and treating disease with advances in machine learning (ML) and artificial intelligence being vitally important in this pursuit. Increasingly, research now utilises DNA methylation measures at cytosine-guanine dinucleotides (CpG) to detect disease and estimate biological traits such as aging. Given the challenge of high dimensionality of DNA methylation data, feature-selection techniques are commonly employed to reduce dimensionality and identify the most important subset of features. In this study, our aim was to test and compare a range of feature-selection methods and ML algorithms in the development of a novel DNA methylation-based telomere length (TL) estimator. We utilised both nested cross-validation and two independent test sets for the comparisons.ResultsWe found that principal component analysis in advance of elastic net regression led to the overall best performing estimator when evaluated using a nested cross-validation analysis and two independent test cohorts. This approach achieved a correlation between estimated and actual TL of 0.295 (83.4% CI [0.201, 0.384]) on the EXTEND test data set. Contrastingly, the baseline model of elastic net regression with no prior feature reduction stage performed less well in general-suggesting a prior feature-selection stage may have important utility. A previously developed TL estimator, DNAmTL, achieved a correlation of 0.216 (83.4% CI [0.118, 0.310]) on the EXTEND data. Additionally, we observed that different DNA methylation-based TL estimators, which have few common CpGs, are associated with many of the same biological entities.ConclusionsThe variance in performance across tested approaches shows that estimators are sensitive to data set heterogeneity and the development of an optimal DNA methylation-based estimator should benefit from the robust methodological approach used in this study. Moreover, our methodology which utilises a range of feature-selection approaches and ML algorithms could be applied to other biological markers and disease phenotypes, to examine their relationship with DNA methylation and predictive value.

Project description:MotivationOne of the major goals in large-scale genomic studies is to identify genes with a prognostic impact on time-to-event outcomes which provide insight into the disease process. With rapid developments in high-throughput genomic technologies in the past two decades, the scientific community is able to monitor the expression levels of tens of thousands of genes and proteins resulting in enormous datasets where the number of genomic features is far greater than the number of subjects. Methods based on univariate Cox regression are often used to select genomic features related to survival outcome; however, the Cox model assumes proportional hazards (PH), which is unlikely to hold for each feature. When applied to genomic features exhibiting some form of non-proportional hazards (NPH), these methods could lead to an under- or over-estimation of the effects. We propose a broad array of marginal screening techniques that aid in feature ranking and selection by accommodating various forms of NPH. First, we develop an approach based on Kullback-Leibler information divergence and the Yang-Prentice model that includes methods for the PH and proportional odds (PO) models as special cases. Next, we propose R2 measures for the PH and PO models that can be interpreted in terms of explained randomness. Lastly, we propose a generalized pseudo-R2 index that includes PH, PO, crossing hazards and crossing odds models as special cases and can be interpreted as the percentage of separability between subjects experiencing the event and not experiencing the event according to feature measurements.ResultsWe evaluate the performance of our measures using extensive simulation studies and publicly available datasets in cancer genomics. We demonstrate that the proposed methods successfully address the issue of NPH in genomic feature selection and outperform existing methods.Availability and implementationR code for the proposed methods is available at github.com/lburns27/Feature-Selection.Contactkarthik.devarajan@fccc.edu.Supplementary informationSupplementary data are available at Bioinformatics online.

Dataset Information

A comparison of feature selection and classification methods in DNA methylation studies using the Illumina Infinium platform.

Publications

A comparison of feature selection and classification methods in DNA methylation studies using the Illumina Infinium platform.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets