simDMRs: Simulate Differentially Methylated Regions

In kdkorthauer/dmrseq: Detection and inference of differentially methylated regions from Whole Genome Bisulfite Sequencing

Simulate Differentially Methylated Regions

Description

Add simulated DMRs to observed control data. Control data will be split into two (artificial) populations.

Usage

simDMRs(bs, num.dmrs = 3000, delta.max0 = 0.3)

Arguments

bs	a BSseq object containing only control samples (from the same population) for which simulated DMRs will be added after dividing the population into two artificial groups.
num.dmrs	an integer specifying how many DMRs to add.
delta.max0	a proportion value indicating the mode value for the difference in proportion of methylated CpGs in the simulated DMRs (the actual value will be drawn from a scaled Beta distribution centered at this value). Default value is 0.3.

Value

A named list object with 5 elements: (1) gr.dmrs is a GRanges object with num.dmrs ranges that represent the random DMRs added. (2) dmr.mncov is a numeric vector that contains the mean coverage in each simulated DMR. (3) dmr.L is a numeric vector that contains the number of CpGs in each simulated DMR. (4) bs is the BSseq object that contains the simulated DMRs. (5) deltas is a numeric vector that contains the effect size used for each DMR.

Examples

# Add simulated DMRs to a BSseq dataset
# This is just for illustrative purposes - ideally you would
# add DMRs to a set of samples from the same condition (in our
# example data, we have data from two different cell types)
# In this case, we shuffle the samples by cell type to create
# a null comparison.

data(BS.chr21)

BS.chr21.sim <- simDMRs(bs=BS.chr21[1:10000,c(1,3,2,4)], 
                        num.dmrs=50)

# show the simulated DMRs GRanges object
show(BS.chr21.sim$gr.dmrs)

# show the updated BSseq object that includes the simulated DMRs
show(BS.chr21.sim$bs)

# examine effect sizes of the DMRs
head(BS.chr21.sim$delta)

kdkorthauer/dmrseq documentation built on Sept. 26, 2024, 9:32 p.m.