# A script to build the annotation files. buildDPkg <- function(pkgName, pkgPath ="/misc/homes/jzhang/madman/datapkgs" ){ sourceURL <- c("www.ncbi.nlm.nih.gov/UniGene", "www.ncbi.nlm.nih.gov/LocusLink", "www.genome.ucsc.edu", "dot.ped.med.umich.edu:200/ourimage/microarrays/Affy_annot/UniGene/index.html", "www.genome.ad.jp/kegg", "www.biostat.harvard.edu/complab/dchip", "www.geneontology.org") library(AnnBuilder) path <- .path.package("AnnBuilder") #define data sources ll <- "ftp://ftp.ncbi.nih.gov/refseq/LocusLink/LL_tmpl.gz" switch(pkgName, "hu6800" = , "hgu133a" = , "hgu95a" = ug <- "ftp://ftp.ncbi.nih.gov/repository/UniGene/Hs.data.gz", "mgu74a" = ug <-"ftp://ftp.ncbi.nih.gov/repository/UniGene/Mm.data.gz", "rgu34a" = ug <-"ftp://ftp.ncbi.nih.gov/repository/UniGene/Rn.data.gz") putIn <- NULL switch(pkgName, "hu6800" = , "hgu133a" = , "hgu95a" = putIn <- "05apr2002", "mgu74a" = putIn <- "mmFeb2002") refLink <- paste("http://genome.cse.ucsc.edu/goldenPath/", putIn, "/database/refLink.txt.gz", sep = "") refGene <- paste("http://genome.cse.ucsc.edu/goldenPath/", putIn, "/database/refGene.txt.gz", sep = "") go <- "ftp://ftp.geneontology.org/pub/go-xml/go_2002-08-termdb.xml.gz" # Download data files from sources llSource <- fileFetcher(ll) ugSource <- fileFetcher(ug) if(!is.null(putIn)){ linkSource <- fileFetcher(refLink) geneSource <- fileFetcher(refGene) } goSource <- fileFetcher(go) # Get the unified mapping between a given id to LocusLink id llNParser <- paste(llSource, "=", file.path(path, "data", "llAffyLLParser"), sep = "") ugNParser <- paste(ugSource, "=", file.path(path, "data", "ugAffyLLParser"), sep = "") if(pkgName == "hgu95a"){ umichFile <- file.path(path, "data", "hgu95av2_UMich") dchipFile <- file.path(path, "data", "hgu95av2_Cheng") tobeMapped <- c(llNParser, ugNParser, umichFile, dchipFile) baseF <- file.path(path, "data", "hgu95av2id") }else if(pkgName == "hu6800"){ umichFile <- file.path(path, "data", "hu6800_UMich") dchipFile <- file.path(path, "data", "hu6800_Cheng") tobeMapped <- c(llNParser, ugNParser, umichFile, dchipFile) baseF <- file.path(path, "data", "hu6800id") }else if(pkgName == "hgu133a"){ umichFile <- file.path(path, "data", "hgu133a_UMich") dchipFile <- file.path(path, "data", "hgu133a_Affy") tobeMapped <- c(llNParser, ugNParser, umichFile, dchipFile) baseF <- file.path(path, "data", "hgu133aid") }else if(pkgName == "mgu74a"){ umichFile <- file.path(path, "data", "mgu74a_UMich") dchipFile <- file.path(path, "data", "mgu74a_Cheng") tobeMapped <- c(llNParser, ugNParser, umichFile, dchipFile) baseF <- file.path(path, "data", "mgu74aid") }else if(pkgName == "rgu34a"){ umichFile <- file.path(path, "data", "rgu34a_UMich") dchipFile <- file.path(path, "data", "rgu34a_Cheng") tobeMapped <- c(llNParser, ugNParser, umichFile, dchipFile) baseF <- file.path(path, "data", "rgu34aid") } testUnified <- file.path(path, "temp","unifiedMapping") acc2LLMapper(outName = testUnified, dbName = "sandBox", tName = "abbuildmap", baseFile = baseF, toMap = tobeMapped, baseCols = c("id", "acc"), byID = "id") # Parse the LocusLink data file using the unified mapping and a base file testLL <- file.path(path, "temp", "buildLL") llParser <- file.path(path, "data", "llParser") fileParser(outName = testLL, baseFile = testUnified, dataFile = llSource, parser = llParser, isDir = FALSE) # Parse chromosomal location and orientation data if(!is.null(putIn)){ testLink <- file.path(path, "temp","buildLink") linkParser <- file.path(path, "data", "refLinkParser") fileParser(outName = testLink, baseFile = testUnified, dataFile = linkSource, parser = linkParser, isDir = FALSE) testGene <- file.path(path, "temp", "buildGene") geneParser <- file.path(path, "data", "refGeneParser") fileParser(outName = testGene, baseFile = testLink, dataFile = geneSource, parser = geneParser, isDir = FALSE) ## Some gene location data may end up with no chromosome number if ## only chromosome number data from LocusLink are used. Get the ## from the refGene file also. geneChr <- file.path(path, "temp", "geneChr") geneParser <- file.path(path, "data", "refGeneChrParser") fileParser(outName = geneChr, baseFile = testLink, dataFile = geneSource, parser = geneParser, isDir = FALSE) # merge the two parsed files f1Col <- c("affy", "acc", "locusid", "unigeneid", "name", "symbol", "chrom", "cyto", "pmid", "grif","sumfun", "go") f2Col <- c("affy", "chrlocation", "chrorientation") testmerged <- file.path(path, "temp", "buildMerged") mergeFiles(file1 = testLL, file2 = testGene, file1Col = f1Col, file2Col = f2Col, idCol = "affy", outName = testmerged, sep = "\t", header = FALSE, isFile = TRUE) }else{ testmerged <- testLL } # Get pathway data and map them to target ids switch(pkgName, "hu6800" = , "hgu133a" = , "hgu95a" = organism <- "human", "mgu74a" = organism <- "mouse", "rgu34a" = organism <- "rat") pathData <- file.path(path, "temp", "humanPath") getPathway(pathData, organism = organism) mappedPath <- file.path(path, "temp", "buildPath") mapPathway(mappedPath, geneFile = testUnified, pathFile = pathData, geneColNames = c("affy", "acc", "ll"), pathColNames = c("ll", "path", "enzyme"), mapId = "ll", colToKeep = c("affy", "path", "enzyme"), geneSep = "\t", pathSep = "\t") # Merge pathway data with previously parsed file if(!is.null(putIn)){ f1Col <- c("affy", "acc", "locusid", "unigeneid", "name", "symbol","chrom", "cyto", "pmid", "grif","sumfun", "go", "chrolocation","chroorientation") }else{ f1Col <- c("affy", "acc", "locusid", "unigeneid", "name", "symbol", "chrom", "cyto", "pmid", "grif","sumfun", "go") } f2Col <- c("affy", "path", "enzyme") mergeFiles(file1 = testmerged, file2 = mappedPath, file1Col = f1Col, file2Col = f2Col, idCol = "affy", outName = testmerged, sep = "\t", header = FALSE, isFile = TRUE) # Produce the XM file (testXML1) containing gene information testXML1 <- file.path(path, "temp", paste(pkgName, ".xml", sep = "")) if(!is.null(putIn)) colNames <- c("AFFY", "ACCNUM", "LOCUSID", "UNIGENE", "GENENAME", "SYMBOL","CHR", "MAP", "PMID", "GRIF", "SUMFUNC", "GO", "CHRLOC", "CHRORI", "PATH", "ENZYME") else colNames <- c("AFFY", "ACCNUM", "LOCUSID", "UNIGENE", "GENENAME", "SYMBOL","CHR", "MAP", "PMID", "GRIF", "SUMFUNC", "GO", "PATH", "ENZYME") multC <- c("PMID", "GO", "CHRLOC", "CHRORI", "PATH", "ENZYME", "CHR","MAP") typeC <- c("GENENAME", "SYMBOL") multS <- ";" typeS <- ";" fileToXML(outName = testXML1, fileName = testmerged, fileCol = colNames, name = pkgName, version = "1.0.1", multCol = multC, typeCol = typeC, multSep = multS, typeSep = typeS, fileSep = "\t", whichOne = "ll", header = FALSE, isFile = TRUE) xml2DataPkg(fileName = testXML1, pkgName = pkgName, urls = sourceURL, path = pkgPath) # Get the reverse mapping between PubMed, pathway, enzyme and Affy ids if(pkgName == "hgu95a"){ hgu95aPMID2AFFY <- file2Env(testmerged, colNames, keyColName = "PMID", valueColName = "AFFY") hgu95aPATH2AFFY <- file2Env(testmerged, colNames, keyColName = "PATH", valueColName = "AFFY") hgu95aENZYME2AFFY <- file2Env(testmerged, colNames, keyColName = "ENZYME", valueColName = "AFFY") assign("hgu95aPMID2AFFY", hgu95aPMID2AFFY, parent.frame()) assign("hgu95aPATH2AFFY", hgu95aPATH2AFFY, parent.frame()) assign("hgu95aENZYME2AFFY", hgu95aENZYME2AFFY, parent.frame()) # if(!is.null(putIn)){ # x <- read.table(testmerged, header = FALSE, as.is = TRUE, # sep = "\t", quote = "", comment.char = "") # y <- read.table(geneChr, header = FALSE, as.is = TRUE, # sep = "\t", quote = "", comment.char = "") # names(x) <- colNames # names(y) <- c("AFFY", "CHR") # hgu95aCHR <- fillCol(x, y, "AFFY", "CHR", "AFFY", "CHR") # environment(hgu95aCHR) <- parent.frame() # makeRdaFile("CHR", pkgName, pkgPath) # } }else if(pkgName == "hu6800"){ hu6800PMID2AFFY <- file2Env(testmerged, colNames, keyColName = "PMID", valueColName = "AFFY") hu6800PATH2AFFY <- file2Env(testmerged, colNames, keyColName = "PATH", valueColName = "AFFY") hu6800ENZYME2AFFY <- file2Env(testmerged, colNames, keyColName = "ENZYME", valueColName = "AFFY") assign("hu6800PMID2AFFY", hu6800PMID2AFFY, parent.frame()) assign("hu6800PATH2AFFY", hu6800PATH2AFFY, parent.frame()) assign("hu6800ENZYME2AFFY", hu6800ENZYME2AFFY, parent.frame()) }else if(pkgName == "hgu133a"){ hgu133aPMID2AFFY <- file2Env(testmerged, colNames, keyColName = "PMID", valueColName = "AFFY") hgu133aPATH2AFFY <- file2Env(testmerged, colNames, keyColName = "PATH", valueColName = "AFFY") hgu133aENZYME2AFFY<- file2Env(testmerged, colNames, keyColName = "ENZYME", valueColName = "AFFY") assign("hgu133aPMID2AFFY", hgu133aPMID2AFFY, parent.frame()) assign("hgu133aPATH2AFFY", hgu133aPATH2AFFY,parent.frame()) assign("hgu133aENZYME2AFFY", hgu133aENZYME2AFFY,parent.frame()) }else if(pkgName == "mgu74a"){ mgu74aPMID2AFFY <- file2Env(testmerged, colNames, keyColName = "PMID", valueColName = "AFFY") mgu74aPATH2AFFY <- file2Env(testmerged, colNames, keyColName = "PATH", valueColName = "AFFY") mgu74aENZYME2AFFY <- file2Env(testmerged, colNames, keyColName = "ENZYME", valueColName = "AFFY") assign("mgu74aPMID2AFFY", mgu74aPMID2AFFY, parent.frame()) assign("mgu74aPATH2AFFY", mgu74aPATH2AFFY, parent.frame()) assign("mgu74aENZYME2AFFY", mgu74aENZYME2AFFY, parent.frame()) }else if(pkgName == "rgu34a"){ rgu34aPMID2AFFY <- file2Env(testmerged, colNames, keyColName = "PMID", valueColName = "AFFY") rgu34aPATH2AFFY <- file2Env(testmerged, colNames, keyColName = "PATH", valueColName = "AFFY") rgu34aENZYME2AFFY <- file2Env(testmerged, colNames, keyColName = "ENZYME", valueColName = "AFFY") assign("rgu34aPMID2AFFY", rgu34aPMID2AFFY, parent.frame()) assign("rgu34aPATH2AFFY", rgu34aPATH2AFFY, parent.frame()) assign("rgu34aENZYME2AFFY", rgu34aENZYME2AFFY, parent.frame()) } makeRdaFile("PMID2AFFY", pkgName, pkgPath, envir = parent.frame()) makeRdaFile("PATH2AFFY", pkgName, pkgPath, envir = parent.frame()) makeRdaFile("ENZYME2AFFY", pkgName, pkgPath, envir = parent.frame()) # Process the GO data file GOXMLParser(dbName = "sandBox", tName = "abtestgoorig", fileName = goSource) goNAffy <- file.path(path, "temp", "testgoNaffy") goParser <- file.path(path, "data", "affyGOParser") fileParser(outName = goNAffy, baseFile=testUnified, dataFile = llSource, parser = goParser, isDir = FALSE) go2GeneMapper(dbName = "sandBox", tName = "abtestgo", fileName = goNAffy, goTName = "abtestgoorig", exclude = "GO:0003673") testXMLByNum <- file.path(path, "temp", paste(pkgName, "ByNum.xml", sep = "")) makeGOByNum(dbName = "sandBox", tName = "abtestgo", outName = testXMLByNum,tColNames = c("go", "geneid","genes","total"), fNames = c("GO","GO2AFFY", "GO2ALLAFFY", "AFFYCOUNTS"),multCol = c("GO2AFFY","GO2ALLAFFY")) xml2DataPkg(fileName = testXMLByNum, pkgName = pkgName, path = pkgPath, rdaOnly = TRUE) #system(paste("cd", file.path(path, "wwwfiles"))) #system("rm *.*") #system(paste("cd", getwd())) } buildGO <- function(pkgName = "GO", pkgPath ="/misc/homes/jzhang/madman/datapkgs"){ dataEnv <- new.env(hash = TRUE, parent = NULL) x <- list(GName = c("molecular_function", "biological_process", "cellular_component"), dbName = "sandBox", tName = "abtestgoorig", idColName = "go", termColName = "ontology", typeColName = "goType", PColName = "parentgoid", topIds = c("GO:0003674","GO:0005575","GO:0008150"), version = "1.0.1") makeGODataPkg(x, "GO", path = pkgPath) }