runMSPC: Assess overlapping enriched regions across multiple sample

In julaiti/MSPC: Multiple Sample Peak Calling: Using combined evidence from replicates to evaluate ChIP-seq peaks

Description

We assess the presence of overlapping enriched regions (A.K.A peaks) across multiple sample simultaneously, repeated evidence over multiple replicates can compensate for lower significance in a single sample, using Fisher method to increase the statistical significance of weak evidence which might get involved in TF or gene regulation.

Usage

runMSPC(peakset, whichType = c("max", "min"),
  replicate.type = c("Biological", "Technical"), cmbStrgThreshold = 1e-08,
  isConfirmed = TRUE)

Arguments

peakset	the output of denoise_ERs. set of Chip-seq replicate imported and all peaks are stored in GRanges object, where all background signal were excluded.
whichType	user has options to keep most stringent(with loest p-value) or least stringent(highest p-value) peak if multiple overlapping peaks were detected. By default, keep most stringent one.
replicate.type	A charcter vector used to select type of input Chip-seq replicate ( Biological / Technical replicate).
cmbStrgThreshold	combined stringency threshold against all enriched regions p-value, so we could identify whether ERs fulfill combined stringency test, and result can be set of confirmed/discarded peaks respectively.
isConfirmed	logical vector that check whether ERs comply combined stringency test.

Value

isConfirmed is True: return set of peaks that both passed from minimum overlapping peak criteria and combined stringency test (Fisher method)

isConfirmed is False : return set of peaks either failed from Fisher method or minimum overlapping requirement.

Note

Special thanks to Martin Morgan's contribution on this revision

Author(s)

Jurat Shahidin

Examples

# set up
library(GenomicRanges)
library(rtracklayer)

# load peak files
files <- getPeakFile()[1:3]
grs <- readPeakFiles(files, pvalueBase=1L)

## Exclude background noise
total.ERs <- denoise_ERs(peakGRs = grs, tau.w = 1.0E-04,
                        overwrite = TRUE)

## explore set of confirmed, discarde peaks
confirmedERs <- runMSPC(peakset = total.ERs, whichType = "max",
                        cmbStrgThreshold = 1.0E-08, isConfirmed = TRUE)
discardedERs <- runMSPC(peakset = total.ERs, whichType = "max",
                        cmbStrgThreshold = 1.0E-08, isConfirmed = FALSE)

julaiti/MSPC documentation built on Oct. 17, 2019, 9:44 p.m.