Skip to main content

Advertisement

Fig. 1 | BMC Bioinformatics

Fig. 1

From: A model of pulldown alignments from SssI-treated DNA improves DNA methylation prediction

Fig. 1

Inputs for running BayMeth. On the left, obtaining read coverage of a genomic window i with some CpG pattern (circles) where the CpG is either methylated (red) or unmethylated (empty). For the experimentally-derived inputs this is done by counting the number of aligned reads that overlap the window from an MBD pulldown experiment done on a sample of interest (yiS=5 for the window depicted on the bottom) or on an SssI-treated sample (yiC=5, for the window depicted in the middle). With our implementation of a calculated SssI Control proxy, we incorporate the range of fragment lengths (, where P() is the probability a fragment of length is in the library) and the amount of SssI pulldown expected (Cn) for a fragment given the number of accessible mCpGs (n) on the fragment. For a given site, x, within our window i, we calculate a term Λx that sums over P()Cn terms for all fragments that begin in the window (on the forward or reverse strands), and then sum over all these Λx values in the window to calculate yiΛ. On the right, arrows indicate which quantities are used as inputs into each BayMeth mode considered

Back to article page