R/outputPeakTable.R

In adductomicsR: Processing of adductomic mass spectral datasets

#'output peak table from AdductQuantif object
#'
#'@param object a 'AdductQuantif' class object
#'@param outputDir character full path to a directory to output the peak to
#'default is the current working directory
#'@return a peaktable with number of rows equal to the number of adducts
#'quantified and 14 peak group information columns plus a number of columns
#'equal to the number of samples quantified.
#'The peak table is saved as a csv file in the output directory
#'named: adductQuantif_peakList_'todays date'.csv.
#'The peak table is also returned to
#'the R session and can be assigned to an object.
#'@usage outputPeakTable(object = NULL, outputDir = NULL)
#'@examples
#'\dontrun{
#'outputPeakTable(object=get(load(paste0(system.file("extdata",
#'package= "adductData"), '/adductQuantResults.Rdata'))))}
#'@export
outputPeakTable <- function(object = NULL, outputDir = NULL) {
    if (is.null(object)) {
        stop("object is missing with no default")
    }
    if (!is(object, "AdductQuantif")) {
        stop("object is not an \"AdductQuantif\" class object")
    }
    if (is.null(outputDir)) {
        outputDir <- getwd()
    }
    outputDir <- paste0(outputDir, "/")
    # output table
    peakQuantTable <- peakQuantTable(object)
    fileNamesTmp <- unique(peakQuantTable[, "file"])
    peakQuantDf <-
        data.frame(matrix(
            peakQuantTable(object)[, "peakArea"],
            byrow = FALSE,
            ncol = length(fileNamesTmp)
        ),
        stringsAsFactors = FALSE)
    colnames(peakQuantDf) <- fileNamesTmp
    # weighted average of column names
    reqPeakColumns <- c(
        "massAcc",
        "rtDev",
        "peakWidth",
        "peakArea",
        "height",
        "mzmed",
        "mzmax",
        "mzmin",
        "corrRtMed",
        "corrRtMin",
        "corrRtMax",
        "rtmed",
        "rtmin",
        "rtmax"
    )
    resMatrixTmp <- matrix(0,
                            nrow = nrow(peakQuantDf),
                            ncol = length(reqPeakColumns))
    colnames(resMatrixTmp) <- reqPeakColumns
    for (i in seq_along(reqPeakColumns)) {
        # change to weighted mean with by
        nonZeroIndx <-
            as.numeric(peakQuantTable[, reqPeakColumns[i]]) != 0
        wMeanColTmp <- tapply(
            as.numeric(peakQuantTable[,reqPeakColumns[i]])[nonZeroIndx],
            as.factor(peakQuantTable[, "featureName"])[nonZeroIndx],
            mean)
        tmpidx <- match(peakQuantTable[seq_len(nrow(resMatrixTmp)),
                                        "featureName"],
                        names(wMeanColTmp))
        resMatrixTmp[, i] <- as.numeric(wMeanColTmp)[tmpidx]
    }
    peakQuantDf <- data.frame(cbind(targTable(object),
                                    resMatrixTmp, peakQuantDf),
                                stringsAsFactors = FALSE)
    write.csv(
        peakQuantDf,
        paste0(outputDir, "adductQuantif_peakList_",
                Sys.Date(),
                ".csv"),
        row.names = FALSE
    )
    return(peakQuantDf)
}  # end function