plotFinalSegmentation: Plotting final segmentation
In MethylSeekR: Segmentation of Bis-seq data

Description Usage Arguments Value Author(s) Examples

This function plots the final segmentations, including PMDs, UMRs and LMRs.

1 2	plotFinalSegmentation(m, segs, PMDs = NA, meth.cutoff, numRegions = 1, pdfFilename = NULL, minCover = 5, nCpG.smoothing = 3)

`m`	GRanges object containing the methylation data.
`segs`	GRanges object containing the UMR/LMR segmentation. Return value of the segmentUMRsLMRs function (see example).
`PMDs`	GRanges object of PMDs. Set to either the return value of the function segmentPMDs (see example) or to NA if there are no PMDs.
`meth.cutoff`	Cut-off on methylation for calling hypomethylated regions.
`numRegions`	The number of randomly chosen regions to be plotted. The default (1) can only be changed if a pdfFilename is provided (see below).
`pdfFilename`	Name of the pdf file in which the figure is saved. If no name is provided (default), the figure is printed to the screen.
`minCover`	Only CpGs with a coverage of at least minCover reads will be used for plotting.
`nCpG.smoothing`	The number of consecutive CpGs that the methylation levels are averaged over.

No return value. The function creates a figure showing the inferred segmentation for a randomly chosen region. The figure is either printed to the screen (default) or saved as a pdf if a filename is provided. If a filename (pdfFilename) is provided, several regions (set via the numRegions argument) can be plotted and saved in a multi-page pdf file. The randomly chosen region that is displayed is broken up into 3 pairs of panels, where in each pair the same region is shown twice, once with raw methylation levels (top) and once with methylation levels smoothed over 3 consecutive CpGs (bottom). In both cases only CpGs with a coverage of at least minCover reads are shown. The raw data best illustrates the disordered nature of methylation levels in PMDs, whereas the smoothed methylation levels more clearly show UMRs and LMRs. In all figures, UMRs are shown as blue squares (placed at the middle of the identified segment), LMRs as red triangles (placed at the middle of the identified segment) and PMDs as green bars (extending over the entire PMD). The cut-off on methylation (meth.cutoff) to determine UMRs and LMRs is shown as a grey dashed line.

Lukas Burger lukas.burger@fmi.ch

library(MethylSeekR)

# get chromosome lengths
library("BSgenome.Hsapiens.UCSC.hg18")
sLengths=seqlengths(Hsapiens)

# read methylation data
methFname <- system.file("extdata", "Lister2009_imr90_hg18_chr22.tab",
package="MethylSeekR")
meth.gr <- readMethylome(FileName=methFname, seqLengths=sLengths)

FDR.cutoff <- 5 
m.sel <- 0.5 
n.sel <- 3

#segment UMRs and LMRs, assuming no PMDs 
UMRLMRsegments.gr <- segmentUMRsLMRs(m=meth.gr, meth.cutoff=m.sel,
nCpG.cutoff=n.sel, myGenomeSeq=Hsapiens, seqLengths=sLengths)

#plot final segmentation, assuming no PMDs
plotFinalSegmentation(m=meth.gr, segs=UMRLMRsegments.gr, meth.cutoff=m.sel, numRegions=1)