Abstract
Abstract Whole-genome bisulfite sequencing (WGBS) provides a precise measure of methylation across the genome, yet pose a challenge on identifying regions that are differentially methylated (DMRs) between different conditions. A number of methods have been proposed, mainly by performing tests on individual methylation locus and combining significant positions. While this approach can globally control locus-level error rates, in the detection of a region, these methods often result in poor estimates. We develop a DMR detecting method MethCP for WGBS data based on change point detection, which naturally segments the genome and provide region-level differential analysis. We demonstrate the performance of MethCP on a senescent cell study, an Arabidopsis dataset and a simulated dataset. We compare our method to four developed methods and we show MethCP is more accurate in detecting the complete DM region.
Availability An Bioconductor package MethCP is upcoming (https://github.com/boyinggong/MethCP).