R/normalizeNpData.R

In cn.farms: cn.FARMS - factor analysis for copy number estimation

#' Processes the non-polymorphic data
#' 
#' Normalization for non-polymorphic data for Affymetrix SNP5 and SNP6
#' 
#' @param filenames the absolute path of the CEL files as a list
#' @param cores number of cores for used for parallelization
#' @param annotDir Optional annotation directory.
#' @param runtype Mode how the results are saved. Possible values are ff or bm. 
#' If ff is chosen the data will not be saved automatically. 
#' With bm the results will be saved permanently. 
#' @param saveFile Name of the file to save.
#' @param method The method for the normalization.
#' @return An instance of \code{\link[Biobase:class.ExpressionSet]{ExpressionSet}} 
#' containing the non-polymorphic data of the microarray.
#' @author Djork-Arne Clevert \email{okko@@clevert.de} and 
#' Andreas Mitterecker \email{mitterecker@@bioinf.jku.at}
#' @examples 
#' \dontrun{
#' library("hapmapsnp6") 
#' celDir <- system.file("celFiles", package = "hapmapsnp6")
#' filenames <- dir(path = celDir, full.names = TRUE)
#' createAnnotation(filenames = filenames)
#' npData <- normalizeNpData(filenames)
#' }
#' @export
normalizeNpData <- function(
        filenames, 
        cores = 1, 
        annotDir = NULL, 
        runtype = "ff", 
        saveFile = "npData", 
        method = c("baseline", "quantiles", "none")) {
    
    method <- match.arg(method)
    normMethods <- c("baseline", "quantiles", "none")
    
    if (!method %in% normMethods) {
        stop("Normalization method not found!")
    }
    
    ## assure correct file extension
    saveFile <- gsub("\\.RData", "", saveFile)
    saveFile <- gsub("\\.rda", "", saveFile)
    saveFile <- paste(saveFile, ".RData", sep = "")
    
    mapping <- affxparser::readCelHeader(filenames[1])$chiptype
    pkgname <- oligo::cleanPlatformName(mapping)
    pkgname <- correctPkgname(pkgname)
    
    if (is.null(annotDir)) {
        vers <- dir(file.path("annotation", pkgname))[1]
        annotDir <- normalizePath(file.path("annotation", pkgname, vers))
    }
    
    if (runtype == "bm" & file.exists(paste(saveFile))) {
        message("Normalization has already been done")
        message("Trying to load normalized data ...")
        load(saveFile)
        message("Done!")
        return(npData)
    }
    
    cat(paste(Sys.time(), "|   Annotation directory: ", 
                    annotDir, " \n"))
    
    featureFile <- file.path(annotDir, "pmfeatureCNV.RData")
    
    if (file.exists(featureFile)) {
        pmfeatureCNV <- data.frame()
        featureSetCNV <- data.frame()
        load(featureFile)    
    } else {
        stop("Something went wrong with the annotation")
    }    
    
    nbrOfSamples <- length(filenames)
    nbrOfProbes <- nrow(pmfeatureCNV)
    
    intensity <- createMatrix(runtype, nbrOfProbes, nbrOfSamples, type = "double",
            bmName = gsub("\\.RData", "", saveFile))
    
    if (cores == 1 & method == "quantiles") {
        tmpExprs <- affxparser::readCelIntensities(filenames, 
                indices = pmfeatureCNV$fid)
        tmpExprs <- log(tmpExprs)
        x <- preprocessCore::normalize.quantiles(tmpExprs)
        intensity[] <- x[]
        rm(x, tmpExprs)
    } else {
        baseline <- determineBaselineArray(
                filenames, nbrOfProbes = 10000, runtype = "ff", 
                pmfeatureCNV$fid, cores = cores)
        baselineArray <- as.vector(affxparser::readCelIntensities(filenames[baseline], 
                indices = pmfeatureCNV$fid))
        
        sfInit(parallel = TRUE, cpus = cores, type = "SOCK")        
        suppressWarnings(sfExport("normalizeAverage", namespace = "cn.farms"))
        suppressWarnings(sfExport("readCelIntensities", namespace = "affxparser"))
        suppressWarnings(sfExport("pmfeatureCNV"))
        suppressWarnings(sfExport("intensity"))
        suppressWarnings(sfExport("filenames"))
        suppressWarnings(sfExport("baselineArray"))    
        
        cat(paste(Sys.time(), "|   Starting processing \n"))
        if (method == "quantiles") {
            suppressWarnings(cnLibrary("preprocessCore", character.only = TRUE))
            res <- suppressWarnings(
                    sfLapply(1:nbrOfSamples, normalizeQuantilesNpH01))
        } else if (method == "none") {
            res <- suppressWarnings(
                    sfLapply(1:nbrOfSamples, normalizeNoneDataH01))     
        } else {
            res <- suppressWarnings(
                    sfLapply(1:nbrOfSamples, normalizeNpDataH01))                
        }
    }
    
    cat(paste(Sys.time(), "|   Non-polymorphic data done \n"))
    
    gc()
    
    sfStop()
    
    featureSetFile <- file.path(annotDir, "featureSetCNV.RData")
    
    if (file.exists(featureSetFile)) {
        load(featureSetFile)    
    } else {
        stop("Something went wrong with the annotation")
    }    
    
    phInf <- featureSetCNV[match(pmfeatureCNV$fsetid, featureSetCNV$fsetid), ]    
    tmpIdx <- which(colnames(phInf) %in% c("chrom_start", "chrom_stop"))
    colnames(phInf)[tmpIdx] <- c("start", "end")
    
    eSet <- new("ExpressionSet")
    assayData(eSet) <- list(intensity = intensity)
    phenoData(eSet) <- new("AnnotatedDataFrame", data = data.frame(filenames))
    featureData(eSet) <- new("AnnotatedDataFrame", data = phInf)
    experimentData(eSet) <- new("MIAME", 
            other = list(
                    annotDir = annotDir, 
                    normalization = "SOR",
                    type = "npData"))    
    annotation(eSet) <- pkgname
    
    sampleNames(eSet) <- gsub("\\.[Cc][Ee][Ll]", "", basename(filenames))    
    
    if (runtype == "bm") {
        cat(paste(Sys.time(), "|   Saving normalized data \n"))
        npData <- eSet
        save(npData, file = saveFile)
    }
    
    return(eSet)
}


#' Helper function 
#' @param i i
#' @return data
#' @author Djork-Arne Clevert \email{okko@@clevert.de} and 
#' Andreas Mitterecker \email{mitterecker@@bioinf.jku.at}
#' @importFrom affxparser readCelIntensities
#' @noRd
normalizeNpDataH01 <- function(i) {
    
    ## non-visible bindings
    filenames <- filenames
    pmfeatureCNV <- pmfeatureCNV
    baselineArray <- baselineArray
    
    LZExprs <- affxparser::readCelIntensities(filenames[i], 
            indices = pmfeatureCNV$fid)
    LZExprs <- normalizeAverage(LZExprs, baselineArray)
    LZExprs <- log2(LZExprs)
    intensity[, i] <- LZExprs
}

#' Helper function 
#' @param i i
#' @return data
#' @author Djork-Arne Clevert \email{okko@@clevert.de} and 
#' Andreas Mitterecker \email{mitterecker@@bioinf.jku.at}
#' @importFrom affxparser readCelIntensities
#' @noRd
normalizeNoneDataH01 <- function(i) {
    
    ## non-visible bindings
    filenames <- filenames
    pmfeatureCNV <- pmfeatureCNV
    baselineArray <- baselineArray
    
    LZExprs <- affxparser::readCelIntensities(filenames[i], 
            indices = pmfeatureCNV$fid)
    #LZExprs <- normalizeAverage(LZExprs, baselineArray)
    LZExprs <- log2(LZExprs)
    intensity[, i] <- LZExprs
}


#' Helper function
#' @param ii ii
#' @param filenames filenames
#' @return Some data
#' @author Djork-Arne Clevert \email{okko@@clevert.de} and 
#' Andreas Mitterecker \email{mitterecker@@bioinf.jku.at}
#' @noRd
normalizeQuantilesNpH01 <- function(i) {
    
    ## non-visible bindings
    filenames <- filenames
    pmfeatureCNV <- pmfeatureCNV
    baselineArray <- baselineArray
    
    tmpExprs <- affxparser::readCelIntensities(filenames[i], 
            indices = pmfeatureCNV$fid)
    intensity[, i] <- log2(preprocessCore::normalize.quantiles.use.target(
                    as.matrix(tmpExprs), baselineArray))
#    intensity[, i] <- log2(baselineArray[rank(tmpExprs)])
}