getBgSubVal: Perform quantification and normalization of count data

In CSSQ: Chip-seq Signal Quantifier Pipeline

Description

This function quantifies each each region for a sample and performs background correction and normalization as instructed. Returns a vector of count information for the input regions.

Usage

getBgSubVal(
  analysisInfo,
  sampleIndex,
  normalizeReadDepth = TRUE,
  normalizeLength = FALSE,
  backgroundSubtract = TRUE,
  countMode = "Union",
  ignore.strand = TRUE,
  inter.feature = FALSE
)

Arguments

analysisInfo	A RangedSummarizedExperiment-class object with regions and sample information from within getRegionCounts.
sampleIndex	Index of the sample to process.
normalizeReadDepth	Logical indicating if count data should be normalized for library sequencing depth. When TRUE (default), counts will be normalized for sequencing depth for each library. When FALSE, no such normalization will be performed and raw counts will be used. (default: TRUE)
normalizeLength	Logical indicating if count data should be normalized to the length of the regions. When TRUE, count data will be normalized for the length of the region being analyzed. When FALSE (default), no such normalization will be performed. (default: FALSE)
backgroundSubtract	Logical indicating if background correction should be performed. When TRUE (default), background subtraction will be performed after length and depth normalization if applicable. When FALSE, no background subtraction will be performed. (default: TRUE)
countMode	Count method passed on to summarizeOverlaps from GenomicAlignments package. ("Union", "IntersectionNotEmpty", "IntersectionStrict" or "user supplied function"). (default: "Union")
ignore.strand	A logical indicating if strand should be considered when matching. (default: TRUE) Passed on to summarizeOverlaps from GenomicAlignments package.
inter.feature	A logical indicating if the 'r countMode' should be aware of overlapping features. When TRUE, reads mapping to multiple features are dropped (i.e., not counted). When FALSE (default), these reads are retained and a count is assigned to each feature they map to. Passed on to summarizeOverlaps from GenomicAlignments package. (default: FALSE)

Value

A vector containing the counts for all the regions.

See Also

getRegionCounts which calls this function

Examples

regionBed <- read.table(system.file("extdata", "chr19_regions.bed", 
package="CSSQ",mustWork = TRUE))
sampleInfo <- read.table(system.file("extdata", "sample_info.txt", 
package="CSSQ",mustWork = TRUE),sep="\t",header=TRUE)
sampleInfo[,3] <- sapply(sampleInfo[,3], 
function(x) system.file("extdata", x, package="CSSQ"))
sampleInfo[,5] <- sapply(sampleInfo[,5], 
function(x) system.file("extdata", x, package="CSSQ"))
regionRange <- GRanges(seqnames=regionBed$V1,
ranges=IRanges(start=regionBed$V2,end=regionBed$V3))
analysisInfo <- SummarizedExperiment(rowRanges=regionRange,
colData=sampleInfo)
NormbgSubCounts <- data.frame(sapply(c(1:nrow(colData(analysisInfo))),
function(x) getBgSubVal(analysisInfo,sampleIndex = x,backgroundSubtract=TRUE,
normalizeReadDepth=TRUE,normalizeLength=FALSE,countMode="Union", 
ignore.strand=TRUE,inter.feature=FALSE)))
NormbgSubCounts

CSSQ documentation built on Nov. 8, 2020, 6:47 p.m.