DMR.plot: Plotting DMRs

In DMRcate: Methylation array and sequencing spatial analysis methods

Description

Plots an individual DMR (in context of possibly other DMRs) as found by dmrcate. Heatmaps are shown as well as proximal coding regions, smoothed group means and chromosome ideogram.

Usage

DMR.plot(ranges, dmr, CpGs, what=c("Beta", "M"), 
         arraytype=c("EPIC", "450K"), phen.col, 
         genome = c("hg19", "hg38", "mm10"), ...)

Arguments

ranges	A GRanges object (ostensibly created by extractRanges()) describing DMR coordinates.
dmr	Index of ranges (one integer only) indicating which DMR to be plotted.
CpGs	Either: - A matrix of beta values for plotting, with unique Illumina probe IDs as rownames. - A GenomicRatioSet, annotated with the appropriate array and data types - A BSseq object containing per-CpG methylation and coverage counts for the samples to be plotted
what	Does CpGs (if a matrix) contain Beta or M-values? Not needed if object is a GenomicRatioSet or BSseq object.
arraytype	Is CpGs (if a matrix) sourced from EPIC or 450K data? Not needed if object is a GenomicRatioSet or BSseq object.
phen.col	Vector of colors denoting phenotypes of all samples described in CpGs. See vignette for worked example.
genome	Reference genome for annotating DMRs. Can be one of "hg19", "hg38" or "mm10"
...	Extra arguments passed to Gviz:::plotTracks().

Value

A plot to the current device.

Author(s)

Aaron Statham <[e-mail]>, Tim J. Peters <[e-mail]>

Examples

library(ExperimentHub)
library(limma)
eh <- ExperimentHub()
FlowSorted.Blood.EPIC <- eh[["EH1136"]]
tcell <- FlowSorted.Blood.EPIC[,colData(FlowSorted.Blood.EPIC)$CD4T==100 |
                                colData(FlowSorted.Blood.EPIC)$CD8T==100]
detP <- detectionP(tcell)
remove <- apply(detP, 1, function (x) any(x > 0.01))
tcell <- tcell[!remove,]
tcell <- preprocessFunnorm(tcell)
#Subset to chr2 only
tcell <- tcell[seqnames(tcell) == "chr2",]
tcellms <- getM(tcell)
tcellms.noSNPs <- rmSNPandCH(tcellms, dist=2, mafcut=0.05)
tcell$Replicate[tcell$Replicate==""] <- tcell$Sample_Name[tcell$Replicate==""]
tcellms.noSNPs <- avearrays(tcellms.noSNPs, tcell$Replicate)
tcell <- tcell[,!duplicated(tcell$Replicate)]
tcell <- tcell[rownames(tcellms.noSNPs),]
colnames(tcellms.noSNPs) <- colnames(tcell)
assays(tcell)[["M"]] <- tcellms.noSNPs
assays(tcell)[["Beta"]] <- ilogit2(tcellms.noSNPs)
type <- factor(tcell$CellType)
design <- model.matrix(~type) 
myannotation <- cpg.annotate("array", tcell, arraytype = "EPIC",
                             analysis.type="differential", design=design, coef=2)
dmrcoutput <- dmrcate(myannotation, lambda=1000, C=2)
results.ranges <- extractRanges(dmrcoutput, genome = "hg19")
groups <- c(CD8T="magenta", CD4T="forestgreen")
cols <- groups[as.character(type)]
DMR.plot(ranges=results.ranges, dmr=1, CpGs=getBeta(tcell), what="Beta", 
         arraytype = "EPIC", phen.col=cols, genome="hg19")

DMRcate documentation built on Jan. 17, 2021, 2 a.m.