regionReads: Count ChIP-Seq Reads for Genomic Regions
In tengmx/ComplexDiff: Differential Binding Estimation for Protein Complexes

Description Usage Arguments Value Examples

Count reads for customized regions or genome wide bins for multiple ChIP-Seq samples in BAM (.bam) or BIGWIG (.bw or .bigwig) format. When no customized peaks provided, this function will generate bin-level counts based on genome information embedded in BAM or BIGWIG files. For BAM files, counts will be generated using Rsubread package; for BIGWIG files, counts will be generated by function in rtracklayer package.

1
2
3

regionReads(samples, colData = DataFrame(cond = basename(samples)),
  peaks = NULL, binsize = 300L, bincut = 0L, readlen = NULL,
  read2pos = "5", ignoreDup = TRUE)

`samples`	A vector of paths to ChIP-seq files in Bam or Bigwig format.
`colData`	A DataFrame object contains at lease one column representing the conditions of samples for comparison in differential binding analysis. This object includes the meta-data describing `samples`. It can store any number of descriptive columns for each sample. See detail in `SummarizedExperiment::colData`. The default here uses sample names to represent conditions.
`peaks`	GRanges or `NULL`. If `NULL`, genome-wide bins with length `binsize` will be pre-selected as potential peaks; otherwise, ChIP-seq peaks should be provided as a GRanges object, e.g. a union of peaks identified by peak caller from all samples. (Default: NULL)
`binsize`	A number specify the bin size. This parameter is only appliable when `peaks` is NULL. (Default: 300)
`bincut`	A numeric cut off on read counts to nominate bins as high coverage bins. This parameter is only appliable when `peaks` is NULL. (Default: 0)
`readlen`	Sample fragment length with which Bigwig files are created. The signal in Bigwig file is `readlen` times of original reads number when created by software as Bedtools. One way to count original reads is to privde `readlen` here. This parameter is only appliable when `samples` are Bigwig files. If `NULL`, it is recommended to multiply `bincut` by a scale factor (normally fragment length). (Default: NULL)
`read2pos`	Parameter for function `featureCounts` in package Rsubread. This parameter is only appliable when `samples` are bam files. (Default: '5')
`ignoreDup`	Parameter for function `featureCounts` in package Rsubread. This parameter is only appliable when `samples` are bam files. (Default: TRUE)

A RangedSummarizedExperiment object containing read counts for genome-wide bins or given peaks..

library(GenomicRanges)
bams <- c(system.file("extdata", "control.bam", package="ComplexDiff"),
          system.file("extdata", "treated.bam", package="ComplexDiff"))
bins <- GRanges("chr1",IRanges(start = seq(1000000,2000000,300),
                      end = seq(1000000,2000000,300)+300-1))
regionReads(samples=bams,peaks=bins)