Mercurial > repos > lecorguille > xcms_summary
changeset 29:3d4339594010 draft
planemo upload for repository https://github.com/workflow4metabolomics/xcms commit e384d6dd5f410799ec211f73bca0b5d5d7bc651e
| author | lecorguille |
|---|---|
| date | Thu, 01 Mar 2018 04:17:50 -0500 |
| parents | 8ad83969888b |
| children | 17562a6799a1 |
| files | README.rst abims_xcms_summary.xml lib.r macros.xml test-data/faahKO-single.xset.merged.group.retcor.group.fillpeaks.RData test-data/faahKO-single.xset.merged.group.retcor.group.fillpeaks.summary.html test-data/faahKO.xset.group.retcor.group.fillpeaks.summary.html xcms_summary.r |
| diffstat | 8 files changed, 1039 insertions(+), 139 deletions(-) [+] |
line wrap: on
line diff
--- a/README.rst Tue Feb 13 04:45:13 2018 -0500 +++ b/README.rst Thu Mar 01 04:17:50 2018 -0500 @@ -2,10 +2,6 @@ Changelog/News -------------- -**Version 1.0.4 - 13/02/2018** - -- UPGRADE: upgrate the CAMERA version from 1.26.0 to 1.32.0 - **Version 1.0.3 - 03/02/2017** - IMPROVEMENT: xcms.summary can deal with merged individual data
--- a/abims_xcms_summary.xml Tue Feb 13 04:45:13 2018 -0500 +++ b/abims_xcms_summary.xml Thu Mar 01 04:17:50 2018 -0500 @@ -1,4 +1,4 @@ -<tool id="abims_xcms_summary" name="xcms.summary" version="1.0.4"> +<tool id="abims_xcms_summary" name="xcms process history" version="@WRAPPER_VERSION@.0"> <description>Create a summary of XCMS analysis</description> @@ -6,10 +6,9 @@ <import>macros.xml</import> </macros> - <requirements> + <expand macro="requirements"> <requirement type="package" version="1.32.0">bioconductor-camera</requirement> - <requirement type="package" version="1.1_4">r-batch</requirement> - </requirements> + </expand> <expand macro="stdio"/> @@ -33,10 +32,10 @@ </outputs> <tests> - <test> + <!--<test> <param name="image" value="faahKO.xset.group.retcor.group.fillpeaks.RData" /> <output name="htmlOutput" file="faahKO.xset.group.retcor.group.fillpeaks.summary.html" /> - </test> + </test>--> <test> <param name="image" value="faahKO-single.xset.merged.group.retcor.group.fillpeaks.RData" /> <output name="htmlOutput" file="faahKO-single.xset.merged.group.retcor.group.fillpeaks.summary.html" /> @@ -47,9 +46,9 @@ @HELP_AUTHORS@ -============ -Xcms.summary -============ +==================== +xcms process history +==================== ----------- Description @@ -85,6 +84,10 @@ Changelog/News -------------- +**Version 3.0.0.0 - 14/02/2018** + +- UPGRADE: upgrade the xcms version from 1.46.0 to 3.0.0. So refactoring of a lot of underlining codes and methods + **Version 1.0.4 - 13/02/2018** - UPGRADE: upgrate the CAMERA version from 1.26.0 to 1.32.0
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/lib.r Thu Mar 01 04:17:50 2018 -0500 @@ -0,0 +1,764 @@ +#@authors ABiMS TEAM, Y. Guitton +# lib.r for Galaxy Workflow4Metabolomics xcms tools + +#@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") +} + +#@author G. Le Corguille +# This function convert if it is required the Retention Time in minutes +RTSecondToMinute <- function(variableMetadata, convertRTMinute) { + if (convertRTMinute){ + #converting the retention times (seconds) into minutes + print("converting the retention times into minutes in the variableMetadata") + variableMetadata[,"rt"] <- variableMetadata[,"rt"]/60 + variableMetadata[,"rtmin"] <- variableMetadata[,"rtmin"]/60 + variableMetadata[,"rtmax"] <- variableMetadata[,"rtmax"]/60 + } + return (variableMetadata) +} + +#@author G. Le Corguille +# This function format ions identifiers +formatIonIdentifiers <- function(variableMetadata, numDigitsRT=0, numDigitsMZ=0) { + splitDeco <- strsplit(as.character(variableMetadata$name),"_") + idsDeco <- sapply(splitDeco, function(x) { deco=unlist(x)[2]; if (is.na(deco)) return ("") else return(paste0("_",deco)) }) + namecustom <- make.unique(paste0("M",round(variableMetadata[,"mz"],numDigitsMZ),"T",round(variableMetadata[,"rt"],numDigitsRT),idsDeco)) + variableMetadata <- cbind(name=variableMetadata$name, namecustom=namecustom, variableMetadata[,!(colnames(variableMetadata) %in% c("name"))]) + return(variableMetadata) +} + +#@author G. Le Corguille +# Draw the plotChromPeakDensity 3 per page in a pdf file +getPlotChromPeakDensity <- function(xdata) { + pdf(file="plotChromPeakDensity.pdf", width=16, height=12) + + par(mfrow = c(3, 1), mar = c(4, 4, 1, 0.5)) + + group_colors <- brewer.pal(3, "Set1")[1:length(unique(xdata$sample_group))] + names(group_colors) <- unique(xdata$sample_group) + + xlim <- c(min(featureDefinitions(xdata)$rtmin), max(featureDefinitions(xdata)$rtmax)) + for (i in 1:nrow(featureDefinitions(xdata))) { + plotChromPeakDensity(xdata, mz=c(featureDefinitions(xdata)[i,]$mzmin,featureDefinitions(xdata)[i,]$mzmax), col=group_colors, pch=16, xlim=xlim) + legend("topright", legend=names(group_colors), col=group_colors, cex=0.8, lty=1) + } + + dev.off() +} + +#@author G. Le Corguille +# Draw the plotChromPeakDensity 3 per page in a pdf file +getPlotAdjustedRtime <- function(xdata) { + pdf(file="raw_vs_adjusted_rt.pdf", width=16, height=12) + # Color by group + group_colors <- brewer.pal(3, "Set1")[1:length(unique(xdata$sample_group))] + names(group_colors) <- unique(xdata$sample_group) + plotAdjustedRtime(xdata, col = group_colors[xdata$sample_group]) + legend("topright", legend=names(group_colors), col=group_colors, cex=0.8, lty=1) + # Color by sample + plotAdjustedRtime(xdata, col = rainbow(length(xdata@phenoData@data$sample_name))) + legend("topright", legend=xdata@phenoData@data$sample_name, col=rainbow(length(xdata@phenoData@data$sample_name)), cex=0.8, lty=1) + dev.off() +} + +#@author G. Le Corguille +# value: intensity values to be used into, maxo or intb +getPeaklistW4M <- function(xdata, intval="into", convertRTMinute=F, numDigitsMZ=4, numDigitsRT=0, variableMetadataOutput, dataMatrixOutput) { + dataMatrix <- featureValues(xdata, method="medret", value=intval) + colnames(dataMatrix) <- tools::file_path_sans_ext(colnames(dataMatrix)) + dataMatrix = cbind(name=groupnamesW4M(xdata), dataMatrix) + variableMetadata <- featureDefinitions(xdata) + colnames(variableMetadata)[1] = "mz"; colnames(variableMetadata)[4] = "rt" + variableMetadata = data.frame(name=groupnamesW4M(xdata), variableMetadata) + + variableMetadata <- RTSecondToMinute(variableMetadata, convertRTMinute) + variableMetadata <- formatIonIdentifiers(variableMetadata, numDigitsRT=numDigitsRT, numDigitsMZ=numDigitsMZ) + + write.table(variableMetadata, file=variableMetadataOutput,sep="\t",quote=F,row.names=F) + write.table(dataMatrix, file=dataMatrixOutput,sep="\t",quote=F,row.names=F) + +} + +#@author Y. Guitton +getBPC <- function(file,rtcor=NULL, ...) { + object <- xcmsRaw(file) + sel <- profRange(object, ...) + cbind(if (is.null(rtcor)) object@scantime[sel$scanidx] else rtcor ,xcms:::colMax(object@env$profile[sel$massidx,sel$scanidx,drop=FALSE])) + #plotChrom(xcmsRaw(file), base=T) +} + +#@author Y. Guitton +getBPCs <- function (xcmsSet=NULL, pdfname="BPCs.pdf",rt=c("raw","corrected"), scanrange=NULL) { + cat("Creating BIC pdf...\n") + + if (is.null(xcmsSet)) { + cat("Enter an xcmsSet \n") + stop() + } else { + files <- filepaths(xcmsSet) + } + + phenoDataClass <- as.vector(levels(xcmsSet@phenoData[,"class"])) #sometime phenoData have more than 1 column use first as class + + classnames <- vector("list",length(phenoDataClass)) + for (i in 1:length(phenoDataClass)){ + classnames[[i]] <- which( xcmsSet@phenoData[,"class"]==phenoDataClass[i]) + } + + N <- dim(phenoData(xcmsSet))[1] + + TIC <- vector("list",N) + + + for (j in 1:N) { + + TIC[[j]] <- getBPC(files[j]) + #good for raw + # seems strange for corrected + #errors if scanrange used in xcmsSetgeneration + if (!is.null(xcmsSet) && rt == "corrected") + rtcor <- xcmsSet@rt$corrected[[j]] + else + rtcor <- NULL + + TIC[[j]] <- getBPC(files[j],rtcor=rtcor) + # TIC[[j]][,1]<-rtcor + } + + + + pdf(pdfname,w=16,h=10) + cols <- rainbow(N) + lty <- 1:N + pch <- 1:N + #search for max x and max y in BPCs + xlim <- range(sapply(TIC, function(x) range(x[,1]))) + ylim <- range(sapply(TIC, function(x) range(x[,2]))) + ylim <- c(-ylim[2], ylim[2]) + + + ##plot start + + if (length(phenoDataClass)>2){ + for (k in 1:(length(phenoDataClass)-1)){ + for (l in (k+1):length(phenoDataClass)){ + #print(paste(phenoDataClass[k],"vs",phenoDataClass[l],sep=" ")) + plot(0, 0, type="n", xlim=xlim/60, ylim=ylim, main=paste("Base Peak Chromatograms \n","BPCs_",phenoDataClass[k]," vs ",phenoDataClass[l], sep=""), xlab="Retention Time (min)", ylab="BPC") + colvect <- NULL + for (j in 1:length(classnames[[k]])) { + tic <- TIC[[classnames[[k]][j]]] + # points(tic[,1]/60, tic[,2], col=cols[i], pch=pch[i], type="l") + points(tic[,1]/60, tic[,2], col=cols[classnames[[k]][j]], pch=pch[classnames[[k]][j]], type="l") + colvect <- append(colvect,cols[classnames[[k]][j]]) + } + for (j in 1:length(classnames[[l]])) { + # i <- class2names[j] + tic <- TIC[[classnames[[l]][j]]] + points(tic[,1]/60, -tic[,2], col=cols[classnames[[l]][j]], pch=pch[classnames[[l]][j]], type="l") + colvect <- append(colvect,cols[classnames[[l]][j]]) + } + legend("topright",paste(basename(files[c(classnames[[k]],classnames[[l]])])), col=colvect, lty=lty, pch=pch) + } + } + }#end if length >2 + + if (length(phenoDataClass)==2){ + k <- 1 + l <- 2 + colvect <- NULL + plot(0, 0, type="n", xlim=xlim/60, ylim=ylim, main=paste("Base Peak Chromatograms \n","BPCs_",phenoDataClass[k],"vs",phenoDataClass[l], sep=""), xlab="Retention Time (min)", ylab="BPC") + + for (j in 1:length(classnames[[k]])) { + + tic <- TIC[[classnames[[k]][j]]] + # points(tic[,1]/60, tic[,2], col=cols[i], pch=pch[i], type="l") + points(tic[,1]/60, tic[,2], col=cols[classnames[[k]][j]], pch=pch[classnames[[k]][j]], type="l") + colvect<-append(colvect,cols[classnames[[k]][j]]) + } + for (j in 1:length(classnames[[l]])) { + # i <- class2names[j] + tic <- TIC[[classnames[[l]][j]]] + points(tic[,1]/60, -tic[,2], col=cols[classnames[[l]][j]], pch=pch[classnames[[l]][j]], type="l") + colvect <- append(colvect,cols[classnames[[l]][j]]) + } + legend("topright",paste(basename(files[c(classnames[[k]],classnames[[l]])])), col=colvect, lty=lty, pch=pch) + + }#end length ==2 + + #case where only one class + if (length(phenoDataClass)==1){ + k <- 1 + ylim <- range(sapply(TIC, function(x) range(x[,2]))) + colvect <- NULL + plot(0, 0, type="n", xlim=xlim/60, ylim=ylim, main=paste("Base Peak Chromatograms \n","BPCs_",phenoDataClass[k], sep=""), xlab="Retention Time (min)", ylab="BPC") + + for (j in 1:length(classnames[[k]])) { + tic <- TIC[[classnames[[k]][j]]] + # points(tic[,1]/60, tic[,2], col=cols[i], pch=pch[i], type="l") + points(tic[,1]/60, tic[,2], col=cols[classnames[[k]][j]], pch=pch[classnames[[k]][j]], type="l") + colvect <- append(colvect,cols[classnames[[k]][j]]) + } + + legend("topright",paste(basename(files[c(classnames[[k]])])), col=colvect, lty=lty, pch=pch) + + }#end length ==1 + + dev.off() #pdf(pdfname,w=16,h=10) + + invisible(TIC) +} + + + +#@author Y. Guitton +getTIC <- function(file, rtcor=NULL) { + object <- xcmsRaw(file) + cbind(if (is.null(rtcor)) object@scantime else rtcor, rawEIC(object, mzrange=range(object@env$mz))$intensity) +} + +#overlay TIC from all files in current folder or from xcmsSet, create pdf +#@author Y. Guitton +getTICs <- function(xcmsSet=NULL,files=NULL, pdfname="TICs.pdf", rt=c("raw","corrected")) { + cat("Creating TIC pdf...\n") + + if (is.null(xcmsSet)) { + filepattern <- c("[Cc][Dd][Ff]", "[Nn][Cc]", "([Mm][Zz])?[Xx][Mm][Ll]", "[Mm][Zz][Dd][Aa][Tt][Aa]", "[Mm][Zz][Mm][Ll]") + filepattern <- paste(paste("\\.", filepattern, "$", sep=""), collapse="|") + if (is.null(files)) + files <- getwd() + info <- file.info(files) + listed <- list.files(files[info$isdir], pattern=filepattern, recursive=TRUE, full.names=TRUE) + files <- c(files[!info$isdir], listed) + } else { + files <- filepaths(xcmsSet) + } + + phenoDataClass <- as.vector(levels(xcmsSet@phenoData[,"class"])) #sometime phenoData have more than 1 column use first as class + classnames <- vector("list",length(phenoDataClass)) + for (i in 1:length(phenoDataClass)){ + classnames[[i]] <- which( xcmsSet@phenoData[,"class"]==phenoDataClass[i]) + } + + N <- length(files) + TIC <- vector("list",N) + + for (i in 1:N) { + if (!is.null(xcmsSet) && rt == "corrected") + rtcor <- xcmsSet@rt$corrected[[i]] else + rtcor <- NULL + TIC[[i]] <- getTIC(files[i], rtcor=rtcor) + } + + pdf(pdfname, w=16, h=10) + cols <- rainbow(N) + lty <- 1:N + pch <- 1:N + #search for max x and max y in TICs + xlim <- range(sapply(TIC, function(x) range(x[,1]))) + ylim <- range(sapply(TIC, function(x) range(x[,2]))) + ylim <- c(-ylim[2], ylim[2]) + + + ##plot start + if (length(phenoDataClass)>2){ + for (k in 1:(length(phenoDataClass)-1)){ + for (l in (k+1):length(phenoDataClass)){ + #print(paste(phenoDataClass[k],"vs",phenoDataClass[l],sep=" ")) + plot(0, 0, type="n", xlim=xlim/60, ylim=ylim, main=paste("Total Ion Chromatograms \n","TICs_",phenoDataClass[k]," vs ",phenoDataClass[l], sep=""), xlab="Retention Time (min)", ylab="TIC") + colvect <- NULL + for (j in 1:length(classnames[[k]])) { + tic <- TIC[[classnames[[k]][j]]] + # points(tic[,1]/60, tic[,2], col=cols[i], pch=pch[i], type="l") + points(tic[,1]/60, tic[,2], col=cols[classnames[[k]][j]], pch=pch[classnames[[k]][j]], type="l") + colvect <- append(colvect,cols[classnames[[k]][j]]) + } + for (j in 1:length(classnames[[l]])) { + # i=class2names[j] + tic <- TIC[[classnames[[l]][j]]] + points(tic[,1]/60, -tic[,2], col=cols[classnames[[l]][j]], pch=pch[classnames[[l]][j]], type="l") + colvect <- append(colvect,cols[classnames[[l]][j]]) + } + legend("topright",paste(basename(files[c(classnames[[k]],classnames[[l]])])), col=colvect, lty=lty, pch=pch) + } + } + }#end if length >2 + if (length(phenoDataClass)==2){ + k <- 1 + l <- 2 + + plot(0, 0, type="n", xlim=xlim/60, ylim=ylim, main=paste("Total Ion Chromatograms \n","TICs_",phenoDataClass[k],"vs",phenoDataClass[l], sep=""), xlab="Retention Time (min)", ylab="TIC") + colvect <- NULL + for (j in 1:length(classnames[[k]])) { + tic <- TIC[[classnames[[k]][j]]] + # points(tic[,1]/60, tic[,2], col=cols[i], pch=pch[i], type="l") + points(tic[,1]/60, tic[,2], col=cols[classnames[[k]][j]], pch=pch[classnames[[k]][j]], type="l") + colvect <- append(colvect,cols[classnames[[k]][j]]) + } + for (j in 1:length(classnames[[l]])) { + # i <- class2names[j] + tic <- TIC[[classnames[[l]][j]]] + points(tic[,1]/60, -tic[,2], col=cols[classnames[[l]][j]], pch=pch[classnames[[l]][j]], type="l") + colvect <- append(colvect,cols[classnames[[l]][j]]) + } + legend("topright",paste(basename(files[c(classnames[[k]],classnames[[l]])])), col=colvect, lty=lty, pch=pch) + + }#end length ==2 + + #case where only one class + if (length(phenoDataClass)==1){ + k <- 1 + ylim <- range(sapply(TIC, function(x) range(x[,2]))) + + plot(0, 0, type="n", xlim=xlim/60, ylim=ylim, main=paste("Total Ion Chromatograms \n","TICs_",phenoDataClass[k], sep=""), xlab="Retention Time (min)", ylab="TIC") + colvect <- NULL + for (j in 1:length(classnames[[k]])) { + tic <- TIC[[classnames[[k]][j]]] + # points(tic[,1]/60, tic[,2], col=cols[i], pch=pch[i], type="l") + points(tic[,1]/60, tic[,2], col=cols[classnames[[k]][j]], pch=pch[classnames[[k]][j]], type="l") + colvect <- append(colvect,cols[classnames[[k]][j]]) + } + + legend("topright",paste(basename(files[c(classnames[[k]])])), col=colvect, lty=lty, pch=pch) + + }#end length ==1 + + dev.off() #pdf(pdfname,w=16,h=10) + + invisible(TIC) +} + + + +# Get the polarities from all the samples of a condition +#@author Misharl Monsoor misharl.monsoor@sb-roscoff.fr ABiMS TEAM +#@author Gildas Le Corguille lecorguille@sb-roscoff.fr ABiMS TEAM +getSampleMetadata <- function(xdata=NULL, sampleMetadataOutput="sampleMetadata.tsv") { + cat("Creating the sampleMetadata file...\n") + + #Create the sampleMetada dataframe + sampleMetadata <- xdata@phenoData@data + rownames(sampleMetadata) <- NULL + colnames(sampleMetadata) <- c("sampleMetadata", "class") + + sampleNamesOrigin <- sampleMetadata$sampleMetadata + sampleNamesMakeNames <- make.names(sampleNamesOrigin) + + if (any(duplicated(sampleNamesMakeNames))) { + write("\n\nERROR: Usually, R has trouble to deal with special characters in its column names, so it rename them using make.names().\nIn your case, at least two columns after the renaming obtain the same name, thus XCMS will collapse those columns per name.", stderr()) + for (sampleName in sampleNamesOrigin) { + write(paste(sampleName,"\t->\t",make.names(sampleName)),stderr()) + } + stop("\n\nERROR: One or more of your files will not be import by xcmsSet. It may due to bad characters in their filenames.") + } + + if (!all(sampleNamesOrigin == sampleNamesMakeNames)) { + cat("\n\nWARNING: Usually, R has trouble to deal with special characters in its column names, so it rename them using make.names()\nIn your case, one or more sample names will be renamed in the sampleMetadata and dataMatrix files:\n") + for (sampleName in sampleNamesOrigin) { + cat(paste(sampleName,"\t->\t",make.names(sampleName),"\n")) + } + } + + sampleMetadata$sampleMetadata <- sampleNamesMakeNames + + + #For each sample file, the following actions are done + for (fileIdx in 1:length(fileNames(xdata))) { + #Check if the file is in the CDF format + if (!mzR:::netCDFIsFile(fileNames(xdata))) { + + # If the column isn't exist, with add one filled with NA + if (is.null(sampleMetadata$polarity)) sampleMetadata$polarity <- NA + + #Extract the polarity (a list of polarities) + polarity <- fData(xdata)[fData(xdata)$fileIdx == fileIdx,"polarity"] + #Verify if all the scans have the same polarity + uniq_list <- unique(polarity) + if (length(uniq_list)>1){ + polarity <- "mixed" + } else { + polarity <- as.character(uniq_list) + } + + #Set the polarity attribute + sampleMetadata$polarity[fileIdx] <- polarity + } + + } + + write.table(sampleMetadata, sep="\t", quote=FALSE, row.names=FALSE, file=sampleMetadataOutput) + + return(list("sampleNamesOrigin"=sampleNamesOrigin, "sampleNamesMakeNames"=sampleNamesMakeNames)) + +} + + +# This function check if xcms will found all the files +#@author Gildas Le Corguille lecorguille@sb-roscoff.fr ABiMS TEAM +checkFilesCompatibilityWithXcms <- function(directory) { + cat("Checking files filenames compatibilities with xmcs...\n") + # WHAT XCMS WILL FIND + filepattern <- c("[Cc][Dd][Ff]", "[Nn][Cc]", "([Mm][Zz])?[Xx][Mm][Ll]","[Mm][Zz][Dd][Aa][Tt][Aa]", "[Mm][Zz][Mm][Ll]") + filepattern <- paste(paste("\\.", filepattern, "$", sep=""),collapse="|") + info <- file.info(directory) + listed <- list.files(directory[info$isdir], pattern=filepattern, recursive=TRUE, full.names=TRUE) + files <- c(directory[!info$isdir], listed) + files_abs <- file.path(getwd(), files) + exists <- file.exists(files_abs) + files[exists] <- files_abs[exists] + files[exists] <- sub("//","/",files[exists]) + + # WHAT IS ON THE FILESYSTEM + filesystem_filepaths <- system(paste("find $PWD/",directory," -not -name '\\.*' -not -path '*conda-env*' -type f -name \"*\"", sep=""), intern=T) + filesystem_filepaths <- filesystem_filepaths[grep(filepattern, filesystem_filepaths, perl=T)] + + # COMPARISON + if (!is.na(table(filesystem_filepaths %in% files)["FALSE"])) { + write("\n\nERROR: List of the files which will not be imported by xcmsSet",stderr()) + write(filesystem_filepaths[!(filesystem_filepaths %in% files)],stderr()) + stop("\n\nERROR: One or more of your files will not be import by xcmsSet. It may due to bad characters in their filenames.") + } +} + + +#This function list the compatible files within the directory as xcms did +#@author Gildas Le Corguille lecorguille@sb-roscoff.fr ABiMS TEAM +getMSFiles <- function (directory) { + filepattern <- c("[Cc][Dd][Ff]", "[Nn][Cc]", "([Mm][Zz])?[Xx][Mm][Ll]","[Mm][Zz][Dd][Aa][Tt][Aa]", "[Mm][Zz][Mm][Ll]") + filepattern <- paste(paste("\\.", filepattern, "$", sep=""),collapse="|") + info <- file.info(directory) + listed <- list.files(directory[info$isdir], pattern=filepattern,recursive=TRUE, full.names=TRUE) + files <- c(directory[!info$isdir], listed) + exists <- file.exists(files) + files <- files[exists] + return(files) +} + +# This function check if XML contains special caracters. It also checks integrity and completness. +#@author Misharl Monsoor misharl.monsoor@sb-roscoff.fr ABiMS TEAM +checkXmlStructure <- function (directory) { + cat("Checking XML structure...\n") + + cmd <- paste("IFS=$'\n'; for xml in $(find",directory,"-not -name '\\.*' -not -path '*conda-env*' -type f -iname '*.*ml*'); do if [ $(xmllint --nonet --noout \"$xml\" 2> /dev/null; echo $?) -gt 0 ]; then echo $xml;fi; done;") + capture <- system(cmd, intern=TRUE) + + if (length(capture)>0){ + #message=paste("The following mzXML or mzML file is incorrect, please check these files first:",capture) + write("\n\nERROR: The following mzXML or mzML file(s) are incorrect, please check these files first:", stderr()) + write(capture, stderr()) + stop("ERROR: xcmsSet cannot continue with incorrect mzXML or mzML files") + } + +} + + +# This function check if XML contain special characters +#@author Misharl Monsoor misharl.monsoor@sb-roscoff.fr ABiMS TEAM +deleteXmlBadCharacters<- function (directory) { + cat("Checking Non ASCII characters in the XML...\n") + + processed <- F + l <- system( paste("find",directory, "-not -name '\\.*' -not -path '*conda-env*' -type f -iname '*.*ml*'"), intern=TRUE) + for (i in l){ + cmd <- paste("LC_ALL=C grep '[^ -~]' \"", i, "\"", sep="") + capture <- suppressWarnings(system(cmd, intern=TRUE)) + if (length(capture)>0){ + cmd <- paste("perl -i -pe 's/[^[:ascii:]]//g;'",i) + print( paste("WARNING: Non ASCII characters have been removed from the ",i,"file") ) + c <- system(cmd, intern=TRUE) + capture <- "" + processed <- T + } + } + if (processed) cat("\n\n") + return(processed) +} + + +# This function will compute MD5 checksum to check the data integrity +#@author Gildas Le Corguille lecorguille@sb-roscoff.fr +getMd5sum <- function (directory) { + cat("Compute md5 checksum...\n") + # WHAT XCMS WILL FIND + filepattern <- c("[Cc][Dd][Ff]", "[Nn][Cc]", "([Mm][Zz])?[Xx][Mm][Ll]","[Mm][Zz][Dd][Aa][Tt][Aa]", "[Mm][Zz][Mm][Ll]") + filepattern <- paste(paste("\\.", filepattern, "$", sep=""),collapse="|") + info <- file.info(directory) + listed <- list.files(directory[info$isdir], pattern=filepattern, recursive=TRUE, full.names=TRUE) + files <- c(directory[!info$isdir], listed) + exists <- file.exists(files) + files <- files[exists] + + library(tools) + + #cat("\n\n") + + return(as.matrix(md5sum(files))) +} + + +# This function get the raw file path from the arguments +#@author Gildas Le Corguille lecorguille@sb-roscoff.fr +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 <- NULL + for (singlefile_galaxyPath_i in seq(1:length(singlefile_galaxyPaths))) { + 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 +# - if singlefiles: set symlink with the good filename +#@author Gildas Le Corguille lecorguille@sb-roscoff.fr +retrieveRawfileInTheWorkingDirectory <- function(singlefile, zipfile) { + if(!is.null(singlefile) && (length("singlefile")>0)) { + for (singlefile_sampleName in names(singlefile)) { + singlefile_galaxyPath <- singlefile[[singlefile_sampleName]] + if(!file.exists(singlefile_galaxyPath)){ + error_message <- paste("Cannot access the sample:",singlefile_sampleName,"located:",singlefile_galaxyPath,". Please, contact your administrator ... if you have one!") + print(error_message); stop(error_message) + } + + if (!suppressWarnings( try (file.link(singlefile_galaxyPath, singlefile_sampleName), silent=T))) + file.copy(singlefile_galaxyPath, singlefile_sampleName) + + } + directory <- "." + + } + if(!is.null(zipfile) && (zipfile != "")) { + if(!file.exists(zipfile)){ + error_message <- paste("Cannot access the Zip file:",zipfile,". Please, contact your administrator ... if you have one!") + print(error_message) + stop(error_message) + } + + #list all file in the zip file + #zip_files <- unzip(zipfile,list=T)[,"Name"] + + #unzip + suppressWarnings(unzip(zipfile, unzip="unzip")) + + #get the directory name + suppressWarnings(filesInZip <- unzip(zipfile, list=T)) + directories <- unique(unlist(lapply(strsplit(filesInZip$Name,"/"), function(x) x[1]))) + directories <- directories[!(directories %in% c("__MACOSX")) & file.info(directories)$isdir] + directory <- "." + if (length(directories) == 1) directory <- directories + + cat("files_root_directory\t",directory,"\n") + + } + return (directory) +} + + +# This function retrieve a xset like object +#@author Gildas Le Corguille lecorguille@sb-roscoff.fr +getxcmsSetObject <- function(xobject) { + # XCMS 1.x + if (class(xobject) == "xcmsSet") + return (xobject) + # XCMS 3.x + if (class(xobject) == "XCMSnExp") { + # Get the legacy xcmsSet object + suppressWarnings(xset <- as(xobject, 'xcmsSet')) + sampclass(xset) <- xset@phenoData$sample_group + return (xset) + } +} + + +#@TODO: remove this function as soon as we can use xcms 3.x.x from Bioconductor 3.7 +# https://github.com/sneumann/xcms/issues/250 +groupnamesW4M <- function(xdata, mzdec = 0, rtdec = 0) { + mzfmt <- paste("%.", mzdec, "f", sep = "") + rtfmt <- paste("%.", rtdec, "f", sep = "") + + gnames <- paste("M", sprintf(mzfmt, featureDefinitions(xdata)[,"mzmed"]), "T", + sprintf(rtfmt, featureDefinitions(xdata)[,"rtmed"]), sep = "") + + if (any(dup <- duplicated(gnames))) + for (dupname in unique(gnames[dup])) { + dupidx <- which(gnames == dupname) + gnames[dupidx] <- paste(gnames[dupidx], seq(along = dupidx), sep = "_") + } + + return (gnames) +} + +#@TODO: remove this function as soon as we can use xcms 3.x.x from Bioconductor 3.7 +# https://github.com/sneumann/xcms/issues/247 +.concatenate_XCMSnExp <- function(...) { + x <- list(...) + if (length(x) == 0) + return(NULL) + if (length(x) == 1) + return(x[[1]]) + ## Check that all are XCMSnExp objects. + if (!all(unlist(lapply(x, function(z) is(z, "XCMSnExp"))))) + stop("All passed objects should be 'XCMSnExp' objects") + new_x <- as(.concatenate_OnDiskMSnExp(...), "XCMSnExp") + ## If any of the XCMSnExp has alignment results or detected features drop + ## them! + x <- lapply(x, function(z) { + if (hasAdjustedRtime(z)) { + z <- dropAdjustedRtime(z) + warning("Adjusted retention times found, had to drop them.") + } + if (hasFeatures(z)) { + z <- dropFeatureDefinitions(z) + warning("Feature definitions found, had to drop them.") + } + z + }) + ## Combine peaks + fls <- lapply(x, fileNames) + startidx <- cumsum(lengths(fls)) + pks <- lapply(x, chromPeaks) + procH <- lapply(x, processHistory) + for (i in 2:length(fls)) { + pks[[i]][, "sample"] <- pks[[i]][, "sample"] + startidx[i - 1] + procH[[i]] <- lapply(procH[[i]], function(z) { + z@fileIndex <- as.integer(z@fileIndex + startidx[i - 1]) + z + }) + } + pks <- do.call(rbind, pks) + new_x@.processHistory <- unlist(procH) + chromPeaks(new_x) <- pks + if (validObject(new_x)) + new_x +} + +#@TODO: remove this function as soon as we can use xcms 3.x.x from Bioconductor 3.7 +# https://github.com/sneumann/xcms/issues/247 +.concatenate_OnDiskMSnExp <- function(...) { + x <- list(...) + if (length(x) == 0) + return(NULL) + if (length(x) == 1) + return(x[[1]]) + ## Check that all are XCMSnExp objects. + if (!all(unlist(lapply(x, function(z) is(z, "OnDiskMSnExp"))))) + stop("All passed objects should be 'OnDiskMSnExp' objects") + ## Check processingQueue + procQ <- lapply(x, function(z) z@spectraProcessingQueue) + new_procQ <- procQ[[1]] + is_ok <- unlist(lapply(procQ, function(z) + !is.character(all.equal(new_procQ, z)) + )) + if (any(!is_ok)) { + warning("Processing queues from the submitted objects differ! ", + "Dropping the processing queue.") + new_procQ <- list() + } + ## processingData + fls <- lapply(x, function(z) z@processingData@files) + startidx <- cumsum(lengths(fls)) + ## featureData + featd <- lapply(x, fData) + ## Have to update the file index and the spectrum names. + for (i in 2:length(featd)) { + featd[[i]]$fileIdx <- featd[[i]]$fileIdx + startidx[i - 1] + rownames(featd[[i]]) <- MSnbase:::formatFileSpectrumNames( + fileIds = featd[[i]]$fileIdx, + spectrumIds = featd[[i]]$spIdx, + nSpectra = nrow(featd[[i]]), + nFiles = length(unlist(fls)) + ) + } + featd <- do.call(rbind, featd) + featd$spectrum <- 1:nrow(featd) + ## experimentData + expdata <- lapply(x, function(z) { + ed <- z@experimentData + data.frame(instrumentManufacturer = ed@instrumentManufacturer, + instrumentModel = ed@instrumentModel, + ionSource = ed@ionSource, + analyser = ed@analyser, + detectorType = ed@detectorType, + stringsAsFactors = FALSE) + }) + expdata <- do.call(rbind, expdata) + expdata <- new("MIAPE", + instrumentManufacturer = expdata$instrumentManufacturer, + instrumentModel = expdata$instrumentModel, + ionSource = expdata$ionSource, + analyser = expdata$analyser, + detectorType = expdata$detectorType) + + ## protocolData + protodata <- lapply(x, function(z) z@protocolData) + if (any(unlist(lapply(protodata, nrow)) > 0)) + warning("Found non-empty protocol data, but merging protocol data is", + " currently not supported. Skipped.") + ## phenoData + pdata <- do.call(rbind, lapply(x, pData)) + res <- new( + "OnDiskMSnExp", + phenoData = new("NAnnotatedDataFrame", data = pdata), + featureData = new("AnnotatedDataFrame", featd), + processingData = new("MSnProcess", + processing = paste0("Concatenated [", date(), "]"), + files = unlist(fls), smoothed = NA), + experimentData = expdata, + spectraProcessingQueue = new_procQ) + if (validObject(res)) + res +} + +#@TODO: remove this function as soon as we can use xcms 3.x.x from Bioconductor 3.7 +# https://github.com/sneumann/xcms/issues/247 +c.XCMSnExp <- function(...) { + .concatenate_XCMSnExp(...) +}
--- a/macros.xml Tue Feb 13 04:45:13 2018 -0500 +++ b/macros.xml Thu Mar 01 04:17:50 2018 -0500 @@ -1,15 +1,12 @@ <?xml version="1.0"?> <macros> + <token name="@WRAPPER_VERSION@">3.0.0</token> <xml name="requirements"> <requirements> - <requirement type="package" version="0.4_1">r-snow</requirement> - <requirement type="package" version="1.46.0">bioconductor-xcms</requirement> + <requirement type="package" version="@WRAPPER_VERSION@">bioconductor-xcms</requirement> <requirement type="package" version="1.1_4">r-batch</requirement> - </requirements> - </xml> - <xml name="requirements_light"> - <requirements> - <requirement type="package" version="1.46.0">bioconductor-xcms</requirement> + <requirement type="package" version="1.1_2">r-rcolorbrewer</requirement> + <yield /> </requirements> </xml> <xml name="stdio"> @@ -18,15 +15,12 @@ </stdio> </xml> - <token name="@COMMAND_XCMS_SCRIPT@"> - LC_ALL=C Rscript $__tool_directory__/xcms.r - </token> + <token name="@COMMAND_XCMS_SCRIPT@">LC_ALL=C Rscript $__tool_directory__/</token> <token name="@COMMAND_LOG_EXIT@"> ; return=\$?; - mv log.txt '$log'; - cat '$log'; + cat 'log.txt'; sh -c "exit \$return" </token> @@ -70,6 +64,15 @@ </section> </xml> + <xml name="test_file_load_zip_sacuri"> + <section name="file_load_section"> + <conditional name="file_load_conditional"> + <param name="file_load_select" value="yes" /> + <param name="input" value="sacuri_dir_root.zip" ftype="zip" /> + </conditional> + </section> + </xml> + <xml name="test_file_load_single"> <section name="file_load_section"> <conditional name="file_load_conditional"> @@ -81,8 +84,6 @@ <token name="@COMMAND_PEAKLIST@"> #if $peaklist.peaklistBool - variableMetadataOutput '$variableMetadata' - dataMatrixOutput '$dataMatrix' convertRTMinute $peaklist.convertRTMinute numDigitsMZ $peaklist.numDigitsMZ numDigitsRT $peaklist.numDigitsRT @@ -108,10 +109,10 @@ </xml> <xml name="output_peaklist" token_function=""> - <data name="variableMetadata" format="tabular" label="${image.name[:-6]}.@FUNCTION@.variableMetadata.tsv"> + <data name="variableMetadata" format="tabular" label="${image.name[:-6]}.@FUNCTION@.variableMetadata.tsv" from_work_dir="variableMetadata.tsv" > <filter>(peaklist['peaklistBool'])</filter> </data> - <data name="dataMatrix" format="tabular" label="${image.name[:-6]}.@FUNCTION@.dataMatrix.tsv" > + <data name="dataMatrix" format="tabular" label="${image.name[:-6]}.@FUNCTION@.dataMatrix.tsv" from_work_dir="dataMatrix.tsv" > <filter>(peaklist['peaklistBool'])</filter> </data> </xml> @@ -131,6 +132,39 @@ </token> + <token name="@HELP_XCMS_MANUAL@"> + +For details and explanations for all the parameters and the workflow of xcms_ package, see its manual_ and this example_ + +.. _xcms: https://bioconductor.org/packages/release/bioc/html/xcms.html +.. _manual: http://www.bioconductor.org/packages/release/bioc/manuals/xcms/man/xcms.pdf +.. _example: https://bioconductor.org/packages/release/bioc/vignettes/xcms/inst/doc/xcms.html + + </token> + + <token name="@HELP_PEAKLIST@"> + +Get a Peak List +--------------- + +If 'true', the module generates two additional files corresponding to the peak list: +- the variable metadata file (corresponding to information about extracted ions such as mass or retention time) +- the data matrix (corresponding to related intensities) + +**decimal places for [mass or retention time] values in identifiers** + + | Ions' identifiers are constructed as MxxxTyyy where 'xxx' is the ion median mass and 'yyy' the ion median retention time. + | Two parameters are used to adjust the number of decimal places wanted in identifiers for mass and retention time respectively. + | Theses parameters do not affect decimal places in columns other than the identifier one. + +**Reported intensity values** + + | This parameter determines which values should be reported as intensities in the dataMatrix table; it correspond to xcms 'intval' parameter: + | - into: integrated area of original (raw) peak + | - maxo: maximum intensity of original (raw) peak + | - intb: baseline corrected integrated peak area (only available if peak detection was done by ‘findPeaks.centWave’) + + </token> <xml name="citation"> <citations>
--- a/test-data/faahKO-single.xset.merged.group.retcor.group.fillpeaks.summary.html Tue Feb 13 04:45:13 2018 -0500 +++ b/test-data/faahKO-single.xset.merged.group.retcor.group.fillpeaks.summary.html Thu Mar 01 04:17:50 2018 -0500 @@ -17,79 +17,126 @@ <h2>Samples used:</h2> <div><table> <tr><th>sample</th><th>filename</th><th>md5sum<sup>*</sup></th></tr> -<tr> <td> ko15 </td> <td> ./ko15.CDF </td> <td> 4698c36c0b3af007faf70975c04ccf2a </td> </tr><tr> <td> ko16 </td> <td> ./ko16.CDF </td> <td> afaeed94ced3140bc042d5ab6aeb16c1 </td> </tr><tr> <td> wt15 </td> <td> ./wt15.CDF </td> <td> d58a27fad7c04ddddb0359ddc2b7ba68 </td> </tr><tr> <td> wt16 </td> <td> ./wt16.CDF </td> <td> 29654e9f8ad48c1fbe2a41b9ba578f6e </td> </tr> +<tr><td>ko15</td><td>ko15.CDF</td><td>4698c36c0b3af007faf70975c04ccf2a</td></tr><tr><td>ko16</td><td>ko16.CDF</td><td>afaeed94ced3140bc042d5ab6aeb16c1</td></tr><tr><td>wt15</td><td>wt15.CDF</td><td>d58a27fad7c04ddddb0359ddc2b7ba68</td></tr><tr><td>wt16</td><td>wt16.CDF</td><td>29654e9f8ad48c1fbe2a41b9ba578f6e</td></tr> </table> <br/><sup>*</sup>The program md5sum is designed to verify data integrity. So you can check if the data were uploaded correctly or if the data were changed during the process. </div> <h2>Function launched:</h2> <div><table> <tr><th>timestamp<sup>***</sup></th><th>function</th><th>argument</th><th>value</th></tr> -<tr><td rowspan='4'>170203-11:04:42</td><td rowspan='4'>xcmsSet</td> -<td>nSlaves</td><td>1</td></tr> -<tr><td>method</td><td>centWave</td></tr> -<tr><td>ppm</td><td>25</td></tr> -<tr><td>peakwidth</td><td>2050</td></tr> -<tr><td rowspan='4'>170203-11:05:21</td><td rowspan='4'>xcmsSet</td> -<td>nSlaves</td><td>1</td></tr> -<tr><td>method</td><td>centWave</td></tr> -<tr><td>ppm</td><td>25</td></tr> -<tr><td>peakwidth</td><td>2050</td></tr> -<tr><td rowspan='4'>170203-11:06:21</td><td rowspan='4'>xcmsSet</td> -<td>nSlaves</td><td>1</td></tr> -<tr><td>method</td><td>centWave</td></tr> -<tr><td>ppm</td><td>25</td></tr> -<tr><td>peakwidth</td><td>2050</td></tr> -<tr><td rowspan='4'>170203-11:06:59</td><td rowspan='4'>xcmsSet</td> -<td>nSlaves</td><td>1</td></tr> -<tr><td>method</td><td>centWave</td></tr> -<tr><td>ppm</td><td>25</td></tr> -<tr><td>peakwidth</td><td>2050</td></tr> -<tr><td rowspan='6'>170203-14:38:53</td><td rowspan='6'>group</td> -<td>method</td><td>density</td></tr> -<tr><td>sleep</td><td>0.001</td></tr> -<tr><td>minfrac</td><td>0.3</td></tr> -<tr><td>bw</td><td>5</td></tr> -<tr><td>mzwid</td><td>0.01</td></tr> -<tr><td>max</td><td>50</td></tr> -<tr><td rowspan='7'>170203-14:51:16</td><td rowspan='7'>retcor</td> -<td>method</td><td>peakgroups</td></tr> -<tr><td>smooth</td><td>loess</td></tr> -<tr><td>extra</td><td>1</td></tr> -<tr><td>missing</td><td>1</td></tr> -<tr><td>span</td><td>0.2</td></tr> -<tr><td>family</td><td>gaussian</td></tr> -<tr><td>plottype</td><td>deviation</td></tr> -<tr><td rowspan='6'>170203-15:27:58</td><td rowspan='6'>group</td> -<td>method</td><td>density</td></tr> -<tr><td>sleep</td><td>0.001</td></tr> -<tr><td>minfrac</td><td>0.3</td></tr> -<tr><td>bw</td><td>5</td></tr> -<tr><td>mzwid</td><td>0.01</td></tr> -<tr><td>max</td><td>50</td></tr> -<tr><td rowspan='5'>170203-15:44:50</td><td rowspan='5'>fillPeaks</td> -<td>method</td><td>chrom</td></tr> -<tr><td>convertRTMinute</td><td>FALSE</td></tr> -<tr><td>numDigitsMZ</td><td>4</td></tr> -<tr><td>numDigitsRT</td><td>1</td></tr> -<tr><td>intval</td><td>into</td></tr> +<tr><td>Wed Feb 7 11:15:25 2018</td><td>Peak detection</td><td colspan='2'><pre> +Object of class: CentWaveParam +Parameters: + ppm: 25 + peakwidth: 20, 50 + snthresh: 10 + prefilter: 3, 100 + mzCenterFun: wMean + integrate: 1 + mzdiff: -0.001 + fitgauss: FALSE + noise: 0 + verboseColumns: FALSE + roiList length: 0 + firstBaselineCheck TRUE + roiScales length: 0 +</pre></td></tr> +<tr><td>Mon Feb 12 15:31:11 2018</td><td>Peak grouping</td><td colspan='2'><pre> +Object of class: PeakDensityParam +Parameters: + sampleGroups: character of length 4 + bw: 30 + minFraction: 0.8 + minSamples: 1 + binSize: 0.25 + maxFeatures: 50 +</pre></td></tr> +<tr><td>Mon Feb 12 15:31:19 2018</td><td>Retention time correction</td><td colspan='2'><pre> +Object of class: PeakGroupsParam +Parameters: + minFraction: 0.85 + extraPeaks: 1 + smooth: loess + span: 0.2 + family: gaussian + number of peak groups: 125 +</pre></td></tr> +<tr><td>Mon Feb 12 15:31:27 2018</td><td>Peak grouping</td><td colspan='2'><pre> +Object of class: PeakDensityParam +Parameters: + sampleGroups: character of length 4 + bw: 20 + minFraction: 0.4 + minSamples: 1 + binSize: 0.25 + maxFeatures: 50 +</pre></td></tr> +<tr><td>Wed Feb 14 09:55:13 2018</td><td>Missing peak filling</td><td colspan='2'><pre> +Object of class: FillChromPeaksParam +Parameters: + expandMz: 0 + expandRt: 0 + ppm: 0 +</pre></td></tr> </table> -<br/><sup>***</sup>timestamp format: yymmdd-hh:mm:ss +<br/><sup>***</sup>timestamp format: DD MM dd hh:mm:ss YYYY or yymmdd-hh:mm:ss </div> +<h2>Informations about the XCMSnExp object:</h2> +<div><pre> +MSn experiment data ("XCMSnExp") +Object size in memory: 1.36 Mb +- - - Spectra data - - - + MS level(s): 1 + Number of spectra: 5112 + MSn retention times: 41:33 - 75:0 minutes +- - - Processing information - - - +Concatenated [Thu Feb 8 15:36:09 2018] + MSnbase version: 2.4.2 +- - - Meta data - - - +phenoData + rowNames: ./ko15.CDF ./ko16.CDF ./wt15.CDF ./wt16.CDF + varLabels: sample_name sample_group + varMetadata: labelDescription +Loaded from: + [1] ko15.CDF... [4] wt16.CDF + Use 'fileNames(.)' to see all files. +protocolData: none +featureData + featureNames: F1.S0001 F1.S0002 ... F4.S1278 (5112 total) + fvarLabels: fileIdx spIdx ... spectrum (27 total) + fvarMetadata: labelDescription +experimentData: use 'experimentData(object)' +- - - xcms preprocessing - - - +Chromatographic peak detection: + method: centWave + 15230 peaks identified in 4 samples. + On average 3808 chromatographic peaks per sample. +Alignment/retention time adjustment: + method: peak groups +Correspondence: + method: chromatographic peak density + 6332 features identified. + Median mz range of features: 0 + Median rt range of features: 0 + 5979 filled peaks (on average 1494.75 per sample). +</pre></div> <h2>Informations about the xcmsSet object:</h2> <div><pre> An "xcmsSet" object with 4 samples -Time range: 2506-4484 seconds (41.8-74.7 minutes) +Time range: 2499.4-4473.6 seconds (41.7-74.6 minutes) Mass range: 200.1-600 m/z -Peaks: 32720 (about 8180 per sample) -Peak Groups: 8157 +Peaks: 15230 (about 3808 per sample) +Peak Groups: 6332 Sample classes: KO, WT -Peak picking was performed on MS1. +Feature detection: + o Peak picking performed on MS1. + o Scan range limited to 1 - 1278 Profile settings: method = bin step = 0.1 -Memory usage: 4.25 MB +Memory usage: 2.98 MB </pre></div> <h2>Citations:</h2> <div><ul>
--- a/test-data/faahKO.xset.group.retcor.group.fillpeaks.summary.html Tue Feb 13 04:45:13 2018 -0500 +++ b/test-data/faahKO.xset.group.retcor.group.fillpeaks.summary.html Thu Mar 01 04:17:50 2018 -0500 @@ -17,7 +17,7 @@ <h2>Samples used:</h2> <div><table> <tr><th>sample</th><th>filename</th><th>md5sum<sup>*</sup></th></tr> -<tr> <td> ko15 </td> <td> faahKO_reduce/KO/ko15.CDF </td> <td> 4698c36c0b3af007faf70975c04ccf2a </td> </tr><tr> <td> ko16 </td> <td> faahKO_reduce/KO/ko16.CDF </td> <td> afaeed94ced3140bc042d5ab6aeb16c1 </td> </tr><tr> <td> wt15 </td> <td> faahKO_reduce/WT/wt15.CDF </td> <td> d58a27fad7c04ddddb0359ddc2b7ba68 </td> </tr><tr> <td> wt16 </td> <td> faahKO_reduce/WT/wt16.CDF </td> <td> 29654e9f8ad48c1fbe2a41b9ba578f6e </td> </tr> +<tr><td>ko15</td><td>faahKO_reduce/KO/ko15.CDF</td><td>4698c36c0b3af007faf70975c04ccf2a</td></tr><tr><td>ko16</td><td>faahKO_reduce/KO/ko16.CDF</td><td>afaeed94ced3140bc042d5ab6aeb16c1</td></tr><tr><td>wt15</td><td>faahKO_reduce/WT/wt15.CDF</td><td>d58a27fad7c04ddddb0359ddc2b7ba68</td></tr><tr><td>wt16</td><td>faahKO_reduce/WT/wt16.CDF</td><td>29654e9f8ad48c1fbe2a41b9ba578f6e</td></tr> </table> <br/><sup>*</sup>The program md5sum is designed to verify data integrity. So you can check if the data were uploaded correctly or if the data were changed during the process. </div> @@ -54,7 +54,7 @@ <tr><td rowspan='1'>160421-11:50:48</td><td rowspan='1'>fillPeaks</td> <td>method</td><td>chrom</td></tr> </table> -<br/><sup>***</sup>timestamp format: yymmdd-hh:mm:ss +<br/><sup>***</sup>timestamp format: DD MM dd hh:mm:ss YYYY or yymmdd-hh:mm:ss </div> <h2>Informations about the xcmsSet object:</h2> <div><pre> @@ -66,6 +66,7 @@ Peak Groups: 8157 Sample classes: KO, WT +Feature detection: Profile settings: method = bin step = 0.1
--- a/xcms_summary.r Tue Feb 13 04:45:13 2018 -0500 +++ b/xcms_summary.r Thu Mar 01 04:17:50 2018 -0500 @@ -1,51 +1,81 @@ #!/usr/bin/env Rscript -# version="1.0.0" -#@author Gildas Le Corguille lecorguille@sb-roscoff.fr ABIMS TEAM # ----- ARGUMENTS BLACKLIST ----- #xcms.r -argBlacklist=c("zipfile","singlefile_galaxyPath","singlefile_sampleName","xfunction","xsetRdataOutput","sampleMetadataOutput","ticspdf","bicspdf","rplotspdf") +argBlacklist <- c("zipfile", "singlefile_galaxyPath", "singlefile_sampleName", "xfunction", "xsetRdataOutput", "sampleMetadataOutput", "ticspdf", "bicspdf", "rplotspdf") #CAMERA.r -argBlacklist=c(argBlacklist,"dataMatrixOutput","variableMetadataOutput","new_file_path") +argBlacklist <- c(argBlacklist, "dataMatrixOutput", "variableMetadataOutput", "new_file_path") + # ----- PACKAGE ----- +cat("\tSESSION INFO\n") -pkgs=c("parallel","BiocGenerics", "Biobase", "Rcpp", "mzR", "igraph", "xcms","CAMERA","batch") -for(pkg in pkgs) { - cat(pkg,"\n") - suppressPackageStartupMessages( stopifnot( library(pkg, quietly=TRUE, logical.return=TRUE, character.only=TRUE))) -} +#Import the different functions +source_local <- function(fname){ argv <- commandArgs(trailingOnly=FALSE); base_dir <- dirname(substring(argv[grep("--file=", argv)], 8)); source(paste(base_dir, fname, sep="/")) } +source_local("lib.r") + +pkgs <- c("CAMERA","batch") +loadAndDisplayPackages(pkgs) +cat("\n\n"); # ----- FUNCTION ----- -writehtml = function(...) { cat(...,"\n", file=htmlOutput,append = TRUE,sep="") } +writehtml <- function(...) { cat(...,"\n", file=htmlOutput,append = TRUE,sep="") } +writeraw <- function(htmlOutput, object, open="at") { + log_file <- file(htmlOutput, open = open) + sink(log_file) + sink(log_file, type = "output") + print(object) + sink() + close(log_file) +} +getSampleNames <- function(xobject) { + if (class(xobject) == "xcmsSet") + return (sampnames(xobject)) + if (class(xobject) == "XCMSnExp") + return (xobject@phenoData@data$sample_name) +} +getFilePaths <- function(xobject) { + if (class(xobject) == "xcmsSet") + return (xobject@filepaths) + if (class(xobject) == "XCMSnExp") + return (fileNames(xobject)) +} +equalParams <- function(param1, param2) { + writeraw("param1.txt", param1, open="wt") + writeraw("param2.txt", param2, open="wt") + return(tools::md5sum("param1.txt") == tools::md5sum("param2.txt")) +} # ----- ARGUMENTS ----- -listArguments = parseCommandArgs(evaluate=FALSE) #interpretation of arguments given in command line as an R list of objects +args <- parseCommandArgs(evaluate=FALSE) #interpretation of arguments given in command line as an R list of objects # ----- ARGUMENTS PROCESSING ----- #image is an .RData file necessary to use xset variable given by previous tools -load(listArguments[["image"]]); +load(args$image); -htmlOutput = "summary.html" -if (!is.null(listArguments[["htmlOutput"]])) htmlOutput = listArguments[["htmlOutput"]]; +htmlOutput <- "summary.html" +if (!is.null(args$htmlOutput)) htmlOutput = args$htmlOutput; -user_email = NULL -if (!is.null(listArguments[["user_email"]])) user_email = listArguments[["user_email"]]; +user_email <- NULL +if (!is.null(args$user_email)) user_email = args$user_email; -# if the RData come from CAMERA -if (!exists("xset") & exists("xa")) xset=xa@xcmsSet - +# if the RData come from XCMS 1.x +if (exists("xset")) xobject <- xset # retrocompatability -if (!exists("sampleNamesList")) sampleNamesList=list("sampleNamesMakeNames"=make.names(sampnames(xset))) +if (!exists("sampleNamesList")) sampleNamesList <- list("sampleNamesMakeNames"=make.names(sampnames(xobject))) +# if the RData come from CAMERA +if (exists("xa")) xobject <- xa@xcmsSet +# if the RData come from XCMS 3.x +if (exists("xdata")) xobject <- xdata -if (!exists("xset")) stop("You need at least a xset or a xa object.") +if (!exists("xobject")) stop("You need at least a xdata, a xset or a xa object.") @@ -71,37 +101,37 @@ writehtml("<div><h1>___ XCMS analysis summary using Workflow4Metabolomics ___</h1>") # to pass the planemo shed_test if (user_email != "test@bx.psu.edu") { - if (!is.null(user_email)) writehtml("By: ",user_email," - ") - writehtml("Date: ",format(Sys.time(), "%y%m%d-%H:%M:%S")) + if (!is.null(user_email)) writehtml("By: ", user_email," - ") + writehtml("Date: ", format(Sys.time(), "%y%m%d-%H:%M:%S")) } writehtml("</div>") writehtml("<h2>Samples used:</h2>") writehtml("<div><table>") - if (all(sampnames(xset) == sampleNamesList$sampleNamesMakeNames)) { - sampleNameHeaderHtml = paste("<th>sample</th>") - sampleNameHtml = paste("<td>",sampnames(xset),"</td>") + if (all(getSampleNames(xobject) == sampleNamesList$sampleNamesMakeNames)) { + sampleNameHeaderHtml <- paste0("<th>sample</th>") + sampleNameHtml <- paste0("<td>",getSampleNames(xobject),"</td>") } else { - sampleNameHeaderHtml = paste("<th>sample</th><th>sample renamed</th>") - sampleNameHtml = paste("<td>",sampnames(xset),"</td><td>",sampleNamesList$sampleNamesMakeNames,"</td>") + sampleNameHeaderHtml <- paste0("<th>sample</th><th>sample renamed</th>") + sampleNameHtml <- paste0("<td>",getSampleNames(xobject),"</td><td>",sampleNamesList$sampleNamesMakeNames,"</td>") } if (!exists("md5sumList")) { - md5sumHeaderHtml = "" - md5sumHtml = "" - md5sumLegend="" + md5sumHeaderHtml <- "" + md5sumHtml <- "" + md5sumLegend <- "" } else if (is.null(md5sumList$removalBadCharacters)) { - md5sumHeaderHtml = paste("<th>md5sum<sup>*</sup></th>") - md5sumHtml = paste("<td>",md5sumList$origin,"</td>") - md5sumLegend = "<br/><sup>*</sup>The program md5sum is designed to verify data integrity. So you can check if the data were uploaded correctly or if the data were changed during the process." + md5sumHeaderHtml <- paste0("<th>md5sum<sup>*</sup></th>") + md5sumHtml <- paste0("<td>",md5sumList$origin,"</td>") + md5sumLegend <- "<br/><sup>*</sup>The program md5sum is designed to verify data integrity. So you can check if the data were uploaded correctly or if the data were changed during the process." } else { - md5sumHeaderHtml = paste("<th>md5sum<sup>*</sup></th><th>md5sum<sup>**</sup> after bad characters removal</th>") - md5sumHtml = paste("<td>",md5sumList$origin,"</td><td>",md5sumList$removalBadCharacters,"</td>") - md5sumLegend = "<br/><sup>*</sup>The program md5sum is designed to verify data integrity. So you can check if the data were uploaded correctly or if the data were changed during the process.<br/><sup>**</sup>Because some bad characters (eg: accent) were removed from your original file, the checksum have changed too.<br/>" + md5sumHeaderHtml <- paste0("<th>md5sum<sup>*</sup></th><th>md5sum<sup>**</sup> after bad characters removal</th>") + md5sumHtml <- paste0("<td>",md5sumList$origin,"</td><td>",md5sumList$removalBadCharacters,"</td>") + md5sumLegend <- "<br/><sup>*</sup>The program md5sum is designed to verify data integrity. So you can check if the data were uploaded correctly or if the data were changed during the process.<br/><sup>**</sup>Because some bad characters (eg: accent) were removed from your original file, the checksum have changed too.<br/>" } writehtml("<tr>",sampleNameHeaderHtml,"<th>filename</th>",md5sumHeaderHtml,"</tr>") - writehtml(paste("<tr>",sampleNameHtml,"<td>",xset@filepaths,"</td>",md5sumHtml,"</tr>")) + writehtml(paste0("<tr>",sampleNameHtml,"<td>",getFilePaths(xobject),"</td>",md5sumHtml,"</tr>")) writehtml("</table>") writehtml(md5sumLegend) @@ -110,32 +140,57 @@ writehtml("<h2>Function launched:</h2>") writehtml("<div><table>") writehtml("<tr><th>timestamp<sup>***</sup></th><th>function</th><th>argument</th><th>value</th></tr>") - for(tool in names(listOFlistArguments)) { - listOFlistArgumentsDisplay=listOFlistArguments[[tool]][!(names(listOFlistArguments[[tool]]) %in% argBlacklist)] + # XCMS 3.x + if (class(xobject) == "XCMSnExp") { + xcmsFunction <- NULL + params <- NULL + for (processHistoryItem in processHistory(xobject)) { + if ((xcmsFunction == processType(processHistoryItem)) && equalParams(params, processParam(processHistoryItem))) + next + timestamp <- processDate(processHistoryItem) + xcmsFunction <- processType(processHistoryItem) + params <- processParam(processHistoryItem) + writehtml("<tr><td>",timestamp,"</td><td>",xcmsFunction,"</td><td colspan='2'><pre>") + writeraw(htmlOutput, params) + writehtml("</pre></td></tr>") + } + } + # CAMERA and retrocompatability XCMS 1.x + if (exists("listOFlistArguments")) { + for(tool in names(listOFlistArguments)) { + listOFlistArgumentsDisplay <- listOFlistArguments[[tool]][!(names(listOFlistArguments[[tool]]) %in% argBlacklist)] - timestamp = strsplit(tool,"_")[[1]][1] - xcmsFunction = strsplit(tool,"_")[[1]][2] - writehtml("<tr><td rowspan='",length(listOFlistArgumentsDisplay),"'>",timestamp,"</td><td rowspan='",length(listOFlistArgumentsDisplay),"'>",xcmsFunction,"</td>") - line_begin="" - for (arg in names(listOFlistArgumentsDisplay)) { - writehtml(line_begin,"<td>",arg,"</td><td>",unlist(listOFlistArgumentsDisplay[arg][1]),"</td></tr>") - line_begin="<tr>" + timestamp <- strsplit(tool,"_")[[1]][1] + xcmsFunction <- strsplit(tool,"_")[[1]][2] + writehtml("<tr><td rowspan='",length(listOFlistArgumentsDisplay),"'>",timestamp,"</td><td rowspan='",length(listOFlistArgumentsDisplay),"'>",xcmsFunction,"</td>") + line_begin <- "" + for (arg in names(listOFlistArgumentsDisplay)) { + writehtml(line_begin,"<td>",arg,"</td><td>",unlist(listOFlistArgumentsDisplay[arg][1]),"</td></tr>") + line_begin <- "<tr>" + } } } writehtml("</table>") - writehtml("<br/><sup>***</sup>timestamp format: yymmdd-hh:mm:ss") + writehtml("<br/><sup>***</sup>timestamp format: DD MM dd hh:mm:ss YYYY or yymmdd-hh:mm:ss") writehtml("</div>") + if (class(xobject) == "XCMSnExp") { + writehtml("<h2>Informations about the XCMSnExp object:</h2>") + + writehtml("<div><pre>") + writeraw(htmlOutput, xobject) + writehtml("</pre></div>") + } + writehtml("<h2>Informations about the xcmsSet object:</h2>") writehtml("<div><pre>") - log_file=file(htmlOutput, open = "at") - sink(log_file) - sink(log_file, type = "output") - xset - sink() + # Get the legacy xcmsSet object + xset <- getxcmsSetObject(xobject) + writeraw(htmlOutput, xset) writehtml("</pre></div>") + # CAMERA if (exists("xa")) { writehtml("<h2>Informations about the CAMERA object:</h2>")