Quantification of DNA Methylation and Epimutations from Bisulfite Sequencing Data – the BEAT package

The following example illustrates a standard use case of the BEAT package, which is used for processing and modeling methylated and unmethylated counts of CG positions from Bisulfite sequencing (BS-Seq) for determination of epimutation rates, i.e. the rate of change in DNA methylation status at CG sites between a reference multi-cell sample and single-cell samples. The input for the package BEAT consists of count data in the form of counts for un-methylated and methylated cytosines per genomic position, which are then grouped into ge-nomic regions of sufficient coverage in order to allow for low-coverage samples to be analyzed. Methylation rates of each region are then modeled using a binomial mixture model in order to adjust for experimental bias, which arises mainly out of incomplete bisulfite conversion (resulting in unmethylated CG positions falsely appearing to be methylated) and sequencing errors (resulting in random errors in methylation status, including methylated CG positions falsely appearing as unmethylated). This vignette explains the use of the package. For a detailed exposition of the statistica