comparison lib.r @ 39:db9bc2c27a0a draft

planemo upload commit d51a0d0a26833671b79711ee18d782e84f301e76

38:2184c0435edf

# lib.r

# This function retrieve a xset like object
#@author Gildas Le Corguille [e-mail]
getxcmsSetObject <- function(xobject) {
    # XCMS 1.x

    variableMetadata=cbind(name=variableMetadata$name, namecustom=namecustom, variableMetadata[,!(colnames(variableMetadata) %in% c("name"))])
    return(variableMetadata)
}

#The function annotateDiffreport without the corr function which bugs
annotatediff <- function(xset=xset, listArguments=listArguments, variableMetadataOutput="variableMetadata.tsv") {
    # Resolve the bug with x11, with the function png
    options(bitmapType='cairo')

    #Check if the fillpeaks step has been done previously, if it hasn't, there is an error message and the execution is stopped.
    res=try(is.null(xset@filled))

    # ------ annot -------
    listArguments[["calcCiS"]]=as.logical(listArguments[["calcCiS"]])
    listArguments[["calcIso"]]=as.logical(listArguments[["calcIso"]])
    listArguments[["calcCaS"]]=as.logical(listArguments[["calcCaS"]])

    # common parameters
    listArguments4annotate = list(object=xset,
    nSlaves=listArguments[["nSlaves"]],sigma=listArguments[["sigma"]],perfwhm=listArguments[["perfwhm"]],
    maxcharge=listArguments[["maxcharge"]],maxiso=listArguments[["maxiso"]],minfrac=listArguments[["minfrac"]],
    ppm=listArguments[["ppm"]],mzabs=listArguments[["mzabs"]],quick=listArguments[["quick"]],
    polarity=listArguments[["polarity"]],max_peaks=listArguments[["max_peaks"]],intval=listArguments[["intval"]])

    # quick == FALSE
    if(listArguments[["quick"]]==FALSE) {
        listArguments4annotate = append(listArguments4annotate,
            list(graphMethod=listArguments[["graphMethod"]],cor_eic_th=listArguments[["cor_eic_th"]],pval=listArguments[["pval"]],
            calcCiS=listArguments[["calcCiS"]],calcIso=listArguments[["calcIso"]],calcCaS=listArguments[["calcCaS"]]))
        # no ruleset
        if (!is.null(listArguments[["multiplier"]])) {
            listArguments4annotate = append(listArguments4annotate,
                list(multiplier=listArguments[["multiplier"]]))
        }
        # ruleset
        else {
            rulset=read.table(listArguments[["rules"]], h=T, sep=";")
            if (ncol(rulset) < 4) rulset=read.table(listArguments[["rules"]], h=T, sep="\t")
            if (ncol(rulset) < 4) rulset=read.table(listArguments[["rules"]], h=T, sep=",")
            if (ncol(rulset) < 4) {
                error_message="Your ruleset file seems not well formatted. The column separators accepted are ; , and tabulation"
                print(error_message)
                stop(error_message)
            }

            listArguments4annotate = append(listArguments4annotate,
                list(rules=rulset))
        }
    }


    # launch annotate
    xa = do.call("annotate", listArguments4annotate)
    peakList=getPeaklist(xa,intval=listArguments[["intval"]])
    peakList=cbind(groupnames(xa@xcmsSet),peakList); colnames(peakList)[1] = c("name");

    # --- Multi condition : diffreport ---
    diffrepOri=NULL
    if (!is.null(listArguments[["runDiffreport"]]) & nlevels(sampclass(xset))>=2) {
        #Check if the fillpeaks step has been done previously, if it hasn't, there is an error message and the execution is stopped.
        res=try(is.null(xset@filled))
        classes=levels(sampclass(xset))
        x=1:(length(classes)-1)
        for (i in seq(along=x) ) {
            y=1:(length(classes))
            for (n in seq(along=y)){
                if(i+n <= length(classes)){
                    filebase=paste(classes[i],class2=classes[i+n],sep="-vs-")

                    diffrep=diffreport(object=xset,class1=classes[i],class2=classes[i+n],filebase=filebase,eicmax=listArguments[["eicmax"]],eicwidth=listArguments[["eicwidth"]],sortpval=TRUE,value=listArguments[["value"]],h=listArguments[["h"]],w=listArguments[["w"]],mzdec=listArguments[["mzdec"]],missing=0)

                    diffrepOri = diffrep

                    # renamming of the column rtmed to rt to fit with camera peaklist function output
                    colnames(diffrep)[colnames(diffrep)=="rtmed"] <- "rt"

                    # combines results and reorder columns
                    diffrep = merge(peakList, diffrep[,c("name","fold","tstat","pvalue")], by.x="name", by.y="name", sort=F)
                    diffrep = cbind(diffrep[,!(colnames(diffrep) %in% c(sampnames(xa@xcmsSet)))],diffrep[,(colnames(diffrep) %in% c(sampnames(xa@xcmsSet)))])

                    diffrep = RTSecondToMinute(diffrep, listArguments[["convertRTMinute"]])
                    diffrep = formatIonIdentifiers(diffrep, numDigitsRT=listArguments[["numDigitsRT"]], numDigitsMZ=listArguments[["numDigitsMZ"]])

                    if(listArguments[["sortpval"]]){
                        diffrep=diffrep[order(diffrep$pvalue), ]
                    }

                    dir.create("tabular", showWarnings = FALSE)
                    write.table(diffrep, sep="\t", quote=FALSE, row.names=FALSE, file=paste("tabular/",filebase,"_tsv.tabular",sep=""))

                    if (listArguments[["eicmax"]] != 0) {
                        if (listArguments[["png2"]] == "pdf")
                            diffreport_png2pdf(filebase)
                    }
                }
            }
        }
        if (listArguments[["png2"]] == "zip")
            diffreport_png2zip()
        if (listArguments[["tabular2"]] == "zip")
            diffreport_tabular2zip()
    }

    # --- variableMetadata ---
    variableMetadata=peakList[,!(make.names(colnames(peakList)) %in% c(make.names(sampnames(xa@xcmsSet))))]
    variableMetadata = RTSecondToMinute(variableMetadata, listArguments[["convertRTMinute"]])
    variableMetadata = formatIonIdentifiers(variableMetadata, numDigitsRT=listArguments[["numDigitsRT"]], numDigitsMZ=listArguments[["numDigitsMZ"]])
    # if we have 2 conditions, we keep stat of diffrep
    if (!is.null(listArguments[["runDiffreport"]]) & nlevels(sampclass(xset))==2) {
        variableMetadata = merge(variableMetadata, diffrep[,c("name","fold","tstat","pvalue")],by.x="name", by.y="name", sort=F)
        if(exists("listArguments[[\"sortpval\"]]")){
            variableMetadata=variableMetadata[order(variableMetadata$pvalue), ]
        }
    }

    variableMetadataOri=variableMetadata

    return(list("xa"=xa,"diffrep"=diffrepOri,"variableMetadata"=variableMetadataOri));

}


combinexsAnnos_function <- function(xaP, xaN, listOFlistArgumentsP,listOFlistArgumentsN, diffrepP=NULL,diffrepN=NULL,pos=TRUE,tol=2,ruleset=NULL,keep_meta=TRUE, convertRTMinute=F, numDigitsMZ=0, numDigitsRT=0, variableMetadataOutput="variableMetadata.tsv"){

    #Load the two Rdata to extract the xset objects from positive and negative mode
    cat("\tObject xset from positive mode\n")
    print(xaP)
    cat("\n")

        stop("You must relauch the CAMERA.annotate step with the lastest version.")
    }

    if(pos){
        xa=xaP
        listOFlistArgumentsP=listOFlistArguments
        mode="neg. Mode"
    } else {
        xa=xaN
        listOFlistArgumentsN=listOFlistArguments
        mode="pos. Mode"
    }

    peakList=getPeaklist(xa)
    peakList=cbind(groupnames(xa@xcmsSet),peakList); colnames(peakList)[1] = c("name");

    return(variableMetadata);

}

# This function get the raw file path from the arguments
getRawfilePathFromArguments <- function(singlefile, zipfile, listArguments) {
    if (!is.null(listArguments[["zipfile"]]))           zipfile = listArguments[["zipfile"]]
    if (!is.null(listArguments[["zipfilePositive"]]))   zipfile = listArguments[["zipfilePositive"]]
    if (!is.null(listArguments[["zipfileNegative"]]))   zipfile = listArguments[["zipfileNegative"]]

    if (!is.null(listArguments[["singlefile_galaxyPath"]])) {
        singlefile_galaxyPaths = listArguments[["singlefile_galaxyPath"]];
        singlefile_sampleNames = listArguments[["singlefile_sampleName"]]
    }
    if (!is.null(listArguments[["singlefile_galaxyPathPositive"]])) {
        singlefile_galaxyPaths = listArguments[["singlefile_galaxyPathPositive"]];
        singlefile_sampleNames = listArguments[["singlefile_sampleNamePositive"]]
    }
    if (!is.null(listArguments[["singlefile_galaxyPathNegative"]])) {
        singlefile_galaxyPaths = listArguments[["singlefile_galaxyPathNegative"]];
        singlefile_sampleNames = listArguments[["singlefile_sampleNameNegative"]]
    }
    if (exists("singlefile_galaxyPaths")){
        singlefile_galaxyPaths = unlist(strsplit(singlefile_galaxyPaths,","))
        singlefile_sampleNames = unlist(strsplit(singlefile_sampleNames,","))

            singlefile_galaxyPath=singlefile_galaxyPaths[singlefile_galaxyPath_i]
            singlefile_sampleName=singlefile_sampleNames[singlefile_galaxyPath_i]
            singlefile[[singlefile_sampleName]] = singlefile_galaxyPath
        }
    }
    for (argument in c("zipfile","zipfilePositive","zipfileNegative","singlefile_galaxyPath","singlefile_sampleName","singlefile_galaxyPathPositive","singlefile_sampleNamePositive","singlefile_galaxyPathNegative","singlefile_sampleNameNegative")) {
        listArguments[[argument]]=NULL
    }
    return(list(zipfile=zipfile, singlefile=singlefile, listArguments=listArguments))
}


# This function retrieve the raw file in the working directory
#   - if zipfile: unzip the file with its directory tree

39:db9bc2c27a0a

# lib.r

#@author G. Le Corguille
# solve an issue with batch if arguments are logical TRUE/FALSE
parseCommandArgs <- function(...) {
    args <- batch::parseCommandArgs(...)
    for (key in names(args)) {
        if (args[key] %in% c("TRUE","FALSE"))
            args[key] = as.logical(args[key])
    }
    return(args)
}

#@author G. Le Corguille
# This function will
# - load the packages
# - display the sessionInfo
loadAndDisplayPackages <- function(pkgs) {
    for(pkg in pkgs) suppressPackageStartupMessages( stopifnot( library(pkg, quietly=TRUE, logical.return=TRUE, character.only=TRUE)))

    sessioninfo = sessionInfo()
    cat(sessioninfo$R.version$version.string,"\n")
    cat("Main packages:\n")
    for (pkg in names(sessioninfo$otherPkgs)) { cat(paste(pkg,packageVersion(pkg)),"\t") }; cat("\n")
    cat("Other loaded packages:\n")
    for (pkg in names(sessioninfo$loadedOnly)) { cat(paste(pkg,packageVersion(pkg)),"\t") }; cat("\n")
}

# This function retrieve a xset like object
#@author Gildas Le Corguille [e-mail]
getxcmsSetObject <- function(xobject) {
    # XCMS 1.x

    variableMetadata=cbind(name=variableMetadata$name, namecustom=namecustom, variableMetadata[,!(colnames(variableMetadata) %in% c("name"))])
    return(variableMetadata)
}

#The function annotateDiffreport without the corr function which bugs
annotatediff <- function(xset=xset, args=args, variableMetadataOutput="variableMetadata.tsv") {
    # Resolve the bug with x11, with the function png
    options(bitmapType='cairo')

    #Check if the fillpeaks step has been done previously, if it hasn't, there is an error message and the execution is stopped.
    res=try(is.null(xset@filled))

    # ------ annot -------
    args$calcCiS=as.logical(args$calcCiS)
    args$calcIso=as.logical(args$calcIso)
    args$calcCaS=as.logical(args$calcCaS)

    # common parameters
    args4annotate = list(object=xset,
        nSlaves=args$nSlaves,sigma=args$sigma,perfwhm=args$perfwhm,
        maxcharge=args$maxcharge,maxiso=args$maxiso,minfrac=args$minfrac,
        ppm=args$ppm,mzabs=args$mzabs,quick=args$quick,
        polarity=args$polarity,max_peaks=args$max_peaks,intval=args$intval)

    # quick == FALSE
    if(args$quick==FALSE) {
        args4annotate = append(args4annotate,
            list(graphMethod=args$graphMethod,cor_eic_th=args$cor_eic_th,pval=args$pval,
            calcCiS=args$calcCiS,calcIso=args$calcIso,calcCaS=args$calcCaS))
        # no ruleset
        if (!is.null(args$multiplier)) {
            args4annotate = append(args4annotate,
                list(multiplier=args$multiplier))
        }
        # ruleset
        else {
            rulset=read.table(args$rules, h=T, sep=";")
            if (ncol(rulset) < 4) rulset=read.table(args$rules, h=T, sep="\t")
            if (ncol(rulset) < 4) rulset=read.table(args$rules, h=T, sep=",")
            if (ncol(rulset) < 4) {
                error_message="Your ruleset file seems not well formatted. The column separators accepted are ; , and tabulation"
                print(error_message)
                stop(error_message)
            }

            args4annotate = append(args4annotate,
                list(rules=rulset))
        }
    }


    # launch annotate
    xa = do.call("annotate", args4annotate)
    peakList=getPeaklist(xa,intval=args$intval)
    peakList=cbind(groupnames(xa@xcmsSet),peakList); colnames(peakList)[1] = c("name");

    # --- Multi condition : diffreport ---
    diffrepOri=NULL
    if (!is.null(args$runDiffreport) & nlevels(sampclass(xset))>=2) {
        #Check if the fillpeaks step has been done previously, if it hasn't, there is an error message and the execution is stopped.
        res=try(is.null(xset@filled))
        classes=levels(sampclass(xset))
        x=1:(length(classes)-1)
        for (i in seq(along=x) ) {
            y=1:(length(classes))
            for (n in seq(along=y)){
                if(i+n <= length(classes)){
                    filebase=paste(classes[i],class2=classes[i+n],sep="-vs-")

                    diffrep=diffreport(
                        object=xset,class1=classes[i],class2=classes[i+n],
                        filebase=filebase,eicmax=args$eicmax,eicwidth=args$eicwidth,
                        sortpval=TRUE,value=args$value,h=args$h,w=args$w,mzdec=args$mzdec,missing=0)

                    diffrepOri = diffrep

                    # renamming of the column rtmed to rt to fit with camera peaklist function output
                    colnames(diffrep)[colnames(diffrep)=="rtmed"] <- "rt"

                    # combines results and reorder columns
                    diffrep = merge(peakList, diffrep[,c("name","fold","tstat","pvalue")], by.x="name", by.y="name", sort=F)
                    diffrep = cbind(diffrep[,!(colnames(diffrep) %in% c(sampnames(xa@xcmsSet)))],diffrep[,(colnames(diffrep) %in% c(sampnames(xa@xcmsSet)))])

                    diffrep = RTSecondToMinute(diffrep, args$convertRTMinute)
                    diffrep = formatIonIdentifiers(diffrep, numDigitsRT=args$numDigitsRT, numDigitsMZ=args$numDigitsMZ)

                    if(args$sortpval){
                        diffrep=diffrep[order(diffrep$pvalue), ]
                    }

                    dir.create("tabular", showWarnings = FALSE)
                    write.table(diffrep, sep="\t", quote=FALSE, row.names=FALSE, file=paste("tabular/",filebase,"_tsv.tabular",sep=""))

                    if (args$eicmax != 0) {
                        if (args$png2 == "pdf")
                            diffreport_png2pdf(filebase)
                    }
                }
            }
        }
        if (args$png2 == "zip")
            diffreport_png2zip()
        if (args$tabular2 == "zip")
            diffreport_tabular2zip()
    }

    # --- variableMetadata ---
    variableMetadata=peakList[,!(make.names(colnames(peakList)) %in% c(make.names(sampnames(xa@xcmsSet))))]
    variableMetadata = RTSecondToMinute(variableMetadata, args$convertRTMinute)
    variableMetadata = formatIonIdentifiers(variableMetadata, numDigitsRT=args$numDigitsRT, numDigitsMZ=args$numDigitsMZ)
    # if we have 2 conditions, we keep stat of diffrep
    if (!is.null(args$runDiffreport) & nlevels(sampclass(xset))==2) {
        variableMetadata = merge(variableMetadata, diffrep[,c("name","fold","tstat","pvalue")],by.x="name", by.y="name", sort=F)
        if(exists("args[[\"sortpval\"]]")){
            variableMetadata=variableMetadata[order(variableMetadata$pvalue), ]
        }
    }

    variableMetadataOri=variableMetadata

    return(list("xa"=xa,"diffrep"=diffrepOri,"variableMetadata"=variableMetadataOri));

}


combinexsAnnos_function <- function(xaP, xaN, diffrepP=NULL,diffrepN=NULL,
    pos=TRUE,tol=2,ruleset=NULL,keep_meta=TRUE, convertRTMinute=F, numDigitsMZ=0,
    numDigitsRT=0, variableMetadataOutput="variableMetadata.tsv"){

    #Load the two Rdata to extract the xset objects from positive and negative mode
    cat("\tObject xset from positive mode\n")
    print(xaP)
    cat("\n")

        stop("You must relauch the CAMERA.annotate step with the lastest version.")
    }

    if(pos){
        xa=xaP
        mode="neg. Mode"
    } else {
        xa=xaN
        mode="pos. Mode"
    }

    peakList=getPeaklist(xa)
    peakList=cbind(groupnames(xa@xcmsSet),peakList); colnames(peakList)[1] = c("name");

    return(variableMetadata);

}

# This function get the raw file path from the arguments
getRawfilePathFromArguments <- function(singlefile, zipfile, args) {
    if (!is.null(args$zipfile))           zipfile = args$zipfile
    if (!is.null(args$zipfilePositive))   zipfile = args$zipfilePositive
    if (!is.null(args$zipfileNegative))   zipfile = args$zipfileNegative

    if (!is.null(args$singlefile_galaxyPath)) {
        singlefile_galaxyPaths = args$singlefile_galaxyPath;
        singlefile_sampleNames = args$singlefile_sampleName
    }
    if (!is.null(args$singlefile_galaxyPathPositive)) {
        singlefile_galaxyPaths = args$singlefile_galaxyPathPositive;
        singlefile_sampleNames = args$singlefile_sampleNamePositive
    }
    if (!is.null(args$singlefile_galaxyPathNegative)) {
        singlefile_galaxyPaths = args$singlefile_galaxyPathNegative;
        singlefile_sampleNames = args$singlefile_sampleNameNegative
    }
    if (exists("singlefile_galaxyPaths")){
        singlefile_galaxyPaths = unlist(strsplit(singlefile_galaxyPaths,","))
        singlefile_sampleNames = unlist(strsplit(singlefile_sampleNames,","))

            singlefile_galaxyPath=singlefile_galaxyPaths[singlefile_galaxyPath_i]
            singlefile_sampleName=singlefile_sampleNames[singlefile_galaxyPath_i]
            singlefile[[singlefile_sampleName]] = singlefile_galaxyPath
        }
    }
    for (argument in c("zipfile", "zipfilePositive", "zipfileNegative",
                        "singlefile_galaxyPath", "singlefile_sampleName",
                        "singlefile_galaxyPathPositive", "singlefile_sampleNamePositive",
                        "singlefile_galaxyPathNegative","singlefile_sampleNameNegative")) {
        args[[argument]]=NULL
    }
    return(list(zipfile=zipfile, singlefile=singlefile, args=args))
}


# This function retrieve the raw file in the working directory
#   - if zipfile: unzip the file with its directory tree