normAndSubtract: Normalization and subtraction of WIG files
In Pasha: Preprocessing of Aligned Sequences from HTS Analyses

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

Function that linearly rescales scores in provided WIG files and subtract Input WIG file if asked

 
normAndSubtractWIG(wigFileList,
                   inputFile=NA, 
                   rescaleInput=FALSE,
                   meanBGLevelRescale=c(0,10), 
                   subtractInput=TRUE, 
                   binSize=50)

`wigFileList`	A character vector, each element describing the full path to the wig file that has to be treated, when several elements, each one will be treated sequentially
`inputFile`	An atomic character vector describing the full path to a wig file from which the genomic scores will eventually be subtracted to wigFileList wig files scores, NA otherwise
`rescaleInput`	An atomic logical deciding if the eventual inputFile scores have to be rescaled (with the same strategy employed for wigFileList files) before using them for subtraction
`meanBGLevelRescale`	A numeric vector of size 2, the first number is the lower limit of the resulting scores after scaling, the second number represent the score that will represent the average enrichment on the genome after rescaling, all scores being rescales linearly around this value
`subtractInput`	An atomic logical specifyijng that the input scores (eventually rescaled) from inputFile, must be subtracted to resulting files or not
`binSize`	An atomic integer value specifying the step size in the wig files

For each wig file in the character vector, this functions reads all the scores for each chromosome, rescale the range of scores, eventually subtract the scores from an input experiment, and saves the results in the same folder as the original file. The rescaling linearly spreads the scores from the lower score to the average score (of the whole original file) to respectively the first and second element of meanBGLevelRescale argument. If an input file is specified, the scores stored for each genomic location are eventually (depending on the arguments) subtracted from corresponding scores in the original wigs. If the user decides to rescale the input before subtraction (recommended), the rescaled inputFile will also be saved in the the original file folder with the _Rescaled extension and can then further be used directly for further computation, with no need to rescale it again (gain of time).

Writes the resulting rescaled and input subtracted files in the same folders as originals, with suffix in filenames _rescaled and _BGsub depending on the parameters

Romain Fenouil

readWIG writeWIG mergeWigs

# Generate artificial wig fixed step files with random counts
exampleFolder <- tempdir() 
nbFiles <- 3
nbChromosomes <- 4

for(fileIndex in 1:nbFiles)
{
    wigScoresContent <- lapply(as.list(1:nbChromosomes), 
        function(x) {sample(x=1:20, size=1000, replace=TRUE, prob=NULL)})
    names(wigScoresContent) <- paste("chr", 1:nbChromosomes, sep="")
    writeWIG(wigScoresContent, 
             paste("wigFileExample", fileIndex, sep=""),
             folder=exampleFolder, 
             fixedStep=200)
}

# Build a file list containing wig files to merge together, and a reference
# to the file to be used as input for score subtraction
wigFileList <- file.path(exampleFolder, 
                         paste("wigFileExample",
                               1:(nbFiles-1), 
                               ".wig", 
                               sep=""))
                                
inputFile <- file.path(exampleFolder, 
                       paste("wigFileExample", 
                             nbFiles,
                             ".wig", 
                             sep=""))

# Run the merge function on 'wigFileList' which will rescales the scores and
# then subtract the wig scores used as input experiment (rescaled too). 
normAndSubtractWIG(wigFileList, 
                   inputFile, 
                   rescaleInput=TRUE,
                   meanBGLevelRescale=c(0,10), 
                   subtractInput=TRUE, 
                   binSize=200)