countanalysis: Determines which regulatory regions are signifigantly altered...

In Mathelab/ALTRE: Workflow for Identifying Regulatory Elements from Chromatin Accessibility Assay Data

Description

Altered regions are those that show differences in chromatin accessibility (using DESeq2 algorithm)

Usage

countanalysis(counts, pval = 0.01, lfcvalue = 1)

Arguments

counts	counts for each region
pval	optional, pvalue considered significant (0.05, 0.01, etc.)
lfcvalue	optional, logfold change value considered significant (value reported on a log scale base 2 so log2fold change of 1.5 means difference in peaks increased by 2^1.5 or 2.8)

Value

list containing: 1) DESeq2 results table 2) some statistics 3) data.frame used for plotting

Examples

## Not run: 
csvfile <- loadCSVFile("DNaseEncodeExample.csv")
samplePeaks <- loadBedFiles(csvfile)
consensusPeaks <- getConsensusPeaks(samplepeaks = samplePeaks, minreps = 2)
TSSannot <- getTSS()
consensusPeaksAnnotated <- combineAnnotatePeaks(conspeaks = consensusPeaks,
   TSS = TSSannot,
   merge = TRUE,
   regionspecific = TRUE,
   distancefromTSSdist = 1500,
   distancefromTSSprox = 1000)
consensusPeaksCounts <- getCounts(annotpeaks = consensusPeaksAnnotated,
   sampleinfo = csvfile,
   reference = 'SAEC',
   chrom = 'chr21')
alteredPeaks <- countanalysis(counts = consensusPeaksCounts,
   pval = 0.01,
   lfcvalue = 1)

## End(Not run)

Mathelab/ALTRE documentation built on May 7, 2019, 3:41 p.m.