findSegments: Find Segments in the provided CN data

In mknoll/cnAnalysis450k: Copy Number Analysis for 450k Illumina Methylation Arrays

Description

Uses data from minfis getCN() function and normalizes probe-wise against control CN data. Segments are identified with changepoints cpr.var() function (BinSeg)

Usage

findSegments(data, ctrl, ctrlAll, statistic = "wilcoxon", plot = FALSE,
  delta = 500, output = "diff", ylim = NULL, arrayType = "auto")

Arguments

data	CN data to evaluate
ctrl	CN data of controls, levels to test to (1 mean / median over all ctrl samples)
ctrlAll	CN data of all control samples
statistic	statistics test to compare groups (controls vs smp); "t.test" or "wilcoxon" (Mann-Whitney-U Test)
plot	plot changepoints
delta	area around changepoints to plot for plot=T
output	"ratio" (Sample/Ctrl) or "diff" (Sample-Ctrl)
ylim	if plot=T give ylim
arrayType	"auto","450k", "EPIC"; auto -> tries to automatically determine the array type (450k, EPIC)

Value

data containing chr, startCG, endCG, segmentmedian, -mean, p-value, SD and samplename

Examples

norm <- minfi::getCN(minfi::preprocessRaw(minfiData::RGsetEx))
ctrlAll <- norm[,5,drop=FALSE]
ctrl <- norm[,4] #ctrl <- apply(ctrlAll, 1, "median")
samples <- norm[,1,drop=FALSE]
findSegments(samples,ctrl, ctrlAll)[1:4,]

mknoll/cnAnalysis450k documentation built on May 23, 2019, 2:01 a.m.