adjustBoundary: Adjust boundaries of peak regions
In mosaics: MOSAiCS (MOdel-based one and two Sample Analysis and Inference for ChIP-Seq)

Description Usage Arguments Details Value Author(s) References See Also Examples

Adjust boundaries of peak regions in the MosaicsPeak class object, which is a peak calling result.

adjustBoundary( object, ... )
## S4 method for signature 'MosaicsPeak'
adjustBoundary( object, minRead=10, extendFromSummit=100,
    trimMinRead1=1.5, trimFC1=5, extendMinRead1=2, extendFC1=50, 
    trimMinRead2=1.5, trimFC2=50, extendMinRead2=1.5, extendFC2=50,
    normC=NA, parallel=FALSE, nCore=8 )

`object`	Object of class `MosaicsPeak`, a peak list object obtained using either functions `mosaicsPeak` or `mosaicsPeakHMM`.
`minRead`	Parameter to determine baseline for trimming and extension of peak boundaries.
`extendFromSummit`	If the updated peak regions do not include the peak summit, either peak start or end is extended by `extendFromSummit`.
`trimMinRead1`	Parameter to determine to trim peak boundaries.
`trimFC1`	Parameter to determine to trim peak boundaries.
`extendMinRead1`	Parameter to determine to extend peak boundaries.
`extendFC1`	Parameter to determine to extend peak boundaries.
`trimMinRead2`	Parameter used to trim peak boundaries.
`trimFC2`	Parameter used to trim peak boundaries.
`extendMinRead2`	Parameter used to extend peak boundaries.
`extendFC2`	Parameter used to extend peak boundaries.
`normC`	Normalizing constant. If not provided, `normC` is estimated as ratio of sequencing depth of ChIP over matched control samples.
`parallel`	Utilize multiple CPUs for parallel computing using `"parallel"` package? Possible values are `TRUE` (utilize multiple CPUs) or `FALSE` (do not utilize multiple CPUs). Default is `FALSE` (do not utilize multiple CPUs).
`nCore`	Number of CPUs when parallel computing is utilized.
`...`	Other parameters to be passed through to generic `mosaicsHMM`.

adjustBoundary adjusts peak boundaries. While adjustBoundary can be applied to a peak list object obtained using either functions mosaicsPeak or mosaicsPeakHMM, adjustBoundary is developed and tested mainly for peak lists from MOSAiCS-HMM model (i.e., from function mosaicsPeakHMM). Note that extractReads should be run first because adjustBoundary is used.

Parallel computing can be utilized for faster computing if parallel=TRUE and parallel package is loaded. nCore determines number of CPUs used for parallel computing.

Construct MosaicsPeak class object.

Dongjun Chung, Pei Fen Kuan, Rene Welch, Sunduz Keles

Kuan, PF, D Chung, G Pan, JA Thomson, R Stewart, and S Keles (2011), "A Statistical Framework for the Analysis of ChIP-Seq Data", Journal of the American Statistical Association, Vol. 106, pp. 891-903.

Chung, D, Zhang Q, and Keles S (2014), "MOSAiCS-HMM: A model-based approach for detecting regions of histone modifications from ChIP-seq data", Datta S and Nettleton D (eds.), Statistical Analysis of Next Generation Sequencing Data, Springer.

mosaicsPeak, mosaicsPeakHMM, extractReads, findSummit, filterPeak, MosaicsPeak.

## Not run: 
library(mosaicsExample)

constructBins( infile=system.file( file.path("extdata","wgEncodeBroadHistoneGm12878H3k4me3StdAlnRep1_chr22_sorted.bam"), package="mosaicsExample"), 
    fileFormat="bam", outfileLoc="./", 
    byChr=FALSE, useChrfile=FALSE, chrfile=NULL, excludeChr=NULL, 
    PET=FALSE, fragLen=200, binSize=200, capping=0 )
constructBins( infile=system.file( file.path("extdata","wgEncodeBroadHistoneGm12878ControlStdAlnRep1_chr22_sorted.bam"), package="mosaicsExample"), 
    fileFormat="bam", outfileLoc="./", 
    byChr=FALSE, useChrfile=FALSE, chrfile=NULL, excludeChr=NULL, 
    PET=FALSE, fragLen=200, binSize=200, capping=0 )

binHM <- readBins( type=c("chip","input"),
    fileName=c( "./wgEncodeBroadHistoneGm12878H3k4me3StdAlnRep1_chr22_sorted.bam_fragL200_bin200.txt",
    "./wgEncodeBroadHistoneGm12878ControlStdAlnRep1_chr22_sorted.bam_fragL200_bin200.txt" ) )
fitHM <- mosaicsFit( binHM, analysisType="IO", bgEst="rMOM" )
hmmHM <- mosaicsFitHMM( fitHM, signalModel = "2S", 
  init="mosaics", init.FDR = 0.05, parallel=TRUE, nCore=8 )
peakHM <- mosaicsPeakHMM( hmmHM, FDR = 0.05, decoding="posterior",
  thres=10, parallel=TRUE, nCore=8 )

peakHM <- extractReads( peakHM,
  chipFile=system.file( file.path("extdata","wgEncodeBroadHistoneGm12878H3k4me3StdAlnRep1_chr22_sorted.bam"), package="mosaicsExample"),
  chipFileFormat="bam", chipPET=FALSE, chipFragLen=200,
  controlFile=system.file( file.path("extdata","wgEncodeBroadHistoneGm12878ControlStdAlnRep1_chr22_sorted.bam"), package="mosaicsExample"), 
  controlFileFormat="bam", controlPET=FALSE, controlFragLen=200, parallel=TRUE, nCore=8 )
peakHM <- findSummit( peakHM, parallel=TRUE, nCore=8 )
peakHM <- adjustBoundary( peakHM, parallel=TRUE, nCore=8 )
peakHM <- filterPeak( peakHM, parallel=TRUE, nCore=8 )

## End(Not run)