Reference-free cell mixture adjustments in analysis of DNA methylation data

This is the publisher’s final pdf. The published article is copyrighted by the author(s) and published by Oxford University Press. The published article can be found at: http://bioinformatics.oxfordjournals.org/.MOTIVATION: Recently there has been increasing interest in the effects\ud of cell mixture on the measurement of DNA methylation, specifically\ud the extent to which small perturbations in cell mixture proportions can\ud register as changes in DNA methylation. A recently published set of\ud statistical methods exploits this association to infer changes in cell\ud mixture proportions, and these methods are presently being applied\ud to adjust for cell mixture effect in the context of epigenome-wide association\ud studies. However, these adjustments require the existence\ud of reference datasets, which may be laborious or expensive to collect.\ud For some tissues such as placenta, saliva, adipose or tumor tissue, the\ud relevant underlying cell types may not be known.\ud RESULTS: We propose a method for conducting epigenome-wide association\ud studies analysis when a reference dataset is unavailable,\ud including a bootstrap method for estimating standard errors. We demonstrate\ud via simulation study and several real data analyses that our\ud proposed method can perform as well as or better than methods that\ud make explicit use of reference datasets. In particular, it may adjust for\ud detailed cell type differences that may be unavailable even in existing\ud reference datasets.\ud AVAILABILITY and IMPLEMENTATION: Software is available in the R\ud package RefFreeEWAS. Data for three of four examples were\ud obtained from Gene Expression Omnibus (GEO), accession numbers\ud GSE37008, GSE42861 and GSE30601, while reference data were\ud obtained from GEO accession number GSE39981