| Previous changeset 7:80216215973c (2014-11-19) Next changeset 9:c28300ae9f69 (2014-11-19) |
|
Commit message:
Uploaded |
|
added:
RNAseqDataAnnotation/RNAseqDataAnnotation.R RNAseqDataAnnotation/RNAseqDataAnnotation.xml RNAseqDataAnnotation/packages.R RNAseqDataAnnotation/tool_dependencies.xml |
| b |
| diff -r 80216215973c -r 73ea91916c9d RNAseqDataAnnotation/RNAseqDataAnnotation.R --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/RNAseqDataAnnotation/RNAseqDataAnnotation.R Wed Nov 19 12:10:09 2014 -0500 |
| [ |
| b'@@ -0,0 +1,209 @@\n+#Author : keime / lornage\n+#Date : 2014/11\n+\n+\n+########################################################################################################\n+#This function concatenates htseq-count result files, normalizes data and annotates data using Ensembl annotations\n+\n+#arguments\n+#path2htseqfiles : path to htseq-count result files\n+#samplenamefile : path ta a tabulated text file with 2 columns : 1. File name 2. Sample names and an header\n+#Species : latin name of the species\n+#ensversion : version of Ensembl to use\n+#fileout : .txt file containing for each library ; gene id, raw read counts, normalized data as well as normalized data/gene length\n+#conversionensembleversion : tab-delimited file allowing conversion of the Ensembl version to the host \n+#\t\t\t\t\t\t\t (Column1 : Version\t Column2 : Host) \n+#conversionensemblname : tab-delimited file allowing conversion of species name to the name of the Ensembl dataset to use\n+#\t\t\t\t\t\t (Column1 : Specie Column2 : Dataset) \n+\n+#output : a data.frame with the following columns :\n+#ensembl gene id\n+#raw read counts for each library (one column per library)\n+#normalized data for each library (one column per library) \n+#normalized data divided by gene length for each library (one column per library)\n+#Gene name\n+#Description\n+\n+#require : biomaRt and DESeq2 Bioconductor packages / package plyr1.8.1\n+\n+#Methods : \n+#Considering that the resulting files of HTSeq-count have 5 lines of comments in the end\n+#Normalization is performed using the method described in Genome Biology 2010;11(10):R106 \n+#and implemented in the DESeq2 Bioconductor package\n+#Gene length correspond to the median of the size of all transcripts corresponding to this gene\n+#########################################################################################################\n+\n+\n+\n+RNAseqDataAnnotation = function(path2htseqfiles, samplenamefile, Species, ensversion, fileout, conversionensemblversion, conversionensemblname){\n+ \t\t\t\t\t\t\t\t\t\t\t\t\n+ #Create a list with the file names in path2htseqfiles \n+\tsampleFiles=list.files(path2htseqfiles)\n+\tsampleFiles=strsplit(sampleFiles,".txt")\n+\t#_noSpikes_htseq\n+\tnfiles=length(sampleFiles) \n+\n+ #Read the data in samplenamefile. Create a data frame establishing the correspondence between file names and sample names\n+\tcorresp = read.table(samplenamefile,header=T,sep="\\t",colClasses=c("character","character"))\n+\tcorresp$File = strsplit(corresp$File,".fastq.gz")\n+\t\n+ #Create a string vector called libnames that contains the name of the samples in the same order as in sampleFiles\n+\tlibnames=rep("",nfiles)\n+\tfor (i in 1:nfiles){\n+\t\tlibnames[i]=corresp$Sample_name[corresp$File==sampleFiles[[i]]]\n+\t}\n+\n+ #For all files located in path2htseqfiles read the corresponding file into R\n+\tlibrary(plyr)\n+\tdatalist = list()\n+\tfor(i in 1:nfiles){\n+\t\trawdata=read.table(paste(paste(path2htseqfiles,sampleFiles[i],sep="/"),"txt",sep="."))\n+\t\t#noSpikes_htseq.\n+\t\tnbrrows=nrow(rawdata)\n+\t\tdatalist[[i]]=rawdata[1:(nbrrows-5), ] # skip the last 5 lines of HTSeq-count files\n+\t\tcolnames(datalist[[i]]) = c("ID",libnames[i])\t\t\n+\t} \n+\t\t\n+ #Join all the files in a data.frame called datafile with rownames = gene id\n+\tdatafile = join_all(datalist, by = "ID", type = "left", match = "all")\n+\t\n+ #Calculate the number of geneID pro file\n+\tnbID=data.frame(rep("",nfiles))\n+\tfor(i in 1:nfiles){\n+\t\tnbID[,i]=nrow(datalist[[i]])\n+\t}\n+\ttotalnbID=apply((nbID[,1:4]),1,sum)\n+\t\n+ #Verify that all the files contain the same gene ID\n+\tif (nrow(datafile)*4==totalnbID[1]){\n+ \n+ #Suppress genes not expressed in all samples \n+\t\tdatafile = datafile[apply(datafile[,2:(nfiles+1)],1,sum)!=0,]\n+\t\trow.names(datafile)=datafile[,1]\n+\t\tdata=datafile[,-1]\n+\t\t\n+ #Number of libraries\n+\t\tnblib= dim(data)[2]\t\n+ #Determine Data + normalization if the specie is not kno'..b'sion<=75){ \n+\t\t\t\tannotation1 = getBM(attributes=c("ensembl_gene_id","external_gene_id","description", "ensembl_transcript_id","exon_chrom_start","exon_chrom_end"),filters="ensembl_gene_id", values=rownames(data), mart=ensembl)\n+\t\t\t}\n+\t\t\telse{\n+\t\t\t\tannotation1 = getBM(attributes=c("ensembl_gene_id","external_gene_name","description", "ensembl_transcript_id","exon_chrom_start","exon_chrom_end"),filters="ensembl_gene_id", values=rownames(data), mart=ensembl)\n+\t\t\t}\t\n+\t\t\t\n+ #because all the annotations are not always found in a first step \n+\t\t\tnot = rownames(data)[!rownames(data) %in% unique(annotation1$ensembl_gene_id)]\n+\t\t\tif (length(not) !=0){\n+\t\t\t\tannotationnot = getBM(attributes=c("ensembl_gene_id","external_gene_id","description", "ensembl_transcript_id","exon_chrom_start","exon_chrom_end"), filters="ensembl_gene_id", values=not, mart=ensembl)\n+\t\t\tannotation = rbind(annotation1, annotationnot)\t\t\n+\t\t\t}\n+\t\t\telse{\n+\t\t\t\tannotation = annotation1\n+\t\t\t}\n+\t\n+ #Exon length\n+\t\t\tensinfos.exlen = data.frame(annotation$ensembl_gene_id, annotation$ensembl_transcript_id, abs(annotation$exon_chrom_start - annotation$exon_chrom_end)+1)\n+\t\t\tcolnames(ensinfos.exlen) = c("ensembl_gene_id", "ensembl_transcript_id", "exon_length")\n+\t\n+ #Transcript length\n+\t\t\ttlen = tapply(ensinfos.exlen$exon_length, ensinfos.exlen$ensembl_transcript_id, sum)\n+\t\t\ttlen.gene = merge(tlen, unique(ensinfos.exlen[,1:2]), by.x="row.names", by.y="ensembl_transcript_id")\n+\t\t\tcolnames(tlen.gene) = c("ensembl_transcript_id", "transcript_length","ensembl_gene_id")\n+\t\n+ #Gene length = median of the size of all transcripts corresponding to this gene\n+\t\t\tglen = tapply(tlen.gene$transcript_length, tlen.gene$ensembl_gene_id, median)\n+\t\n+ #Data with gene length\n+\t\t\tdatalen = merge(data, glen, by="row.names") \n+\t\t\tcolnames(datalen) = c("Ensembl_gene_id",colnames(data), "Gene_length")\n+\t\n+ #Data with annotations and gene length\n+\t\t\tannotationgene = unique(annotation[,1:3])\n+\t\t\tdataannot = merge(datalen, annotationgene, by.x="Ensembl_gene_id", by.y="ensembl_gene_id")\n+\t\n+ #To keep only the first part of the gene description (before [)\n+\t\t\ttmpdesc = strsplit(as.character(dataannot$description),"[", fixed=T)\n+\t\t\tf = function(l){\n+\t\t\t\tif (length(l)>=1){\n+\t\t\t\t\treturn(l[[1]])\n+\t\t\t\t}\n+\t\t\t\telse{\n+\t\t\t\t\treturn("")\n+\t\t\t\t}\n+\t\t\t}\n+\t\t\ttmpdescok = unlist(lapply(tmpdesc, f))\n+\t\t\tdataannot$description = tmpdescok\n+\t\n+ #Normalized data calculation\n+\t\t\tnbcol = dim(dataannot)[2] #nb of column in the data.frame\n+\t\t\tlibrary(DESeq2)\n+\t\t\tconds = factor(1:nblib)\n+\t\t\tdesign = data.frame(Condition=conds)\n+\t\t\tdds = DESeqDataSetFromMatrix(countData=dataannot[,-c(1,nbcol,nbcol-1,nbcol-2)], colData=design, design=~Condition)\n+\t\t\tdds = estimateSizeFactors(dds)\n+\t\t\tdatanorm = t(t(dataannot[,-c(1,nbcol,nbcol-1,nbcol-2)])/sizeFactors(dds))\n+\t\n+ #Normalized data adjusted for gene length (normalized data / gene length)\n+\t\t\trpkn = datanorm / (as.vector(dataannot[,nbcol-2]/1000 ))\n+\t\n+ #Data + annotations + rpkn\n+\t\t\tdataall = data.frame(dataannot[,-c(nbcol,nbcol-1,nbcol-2)] , datanorm, rpkn, dataannot[,c(nbcol-1,nbcol)] )\n+\t\t\n+ #Renames columns\n+\t\t\tcolnames(dataall) = c("Ensembl gene id", paste(libnames,"(raw read counts)"), paste(libnames,"(normalized)"), paste(libnames,"(normalized and divided by gene length in kb)"), "Gene name", "Description")\n+\t\t\twrite.table(dataall, file=fileout, sep="\\t", quote=F, row.names=F)\n+\n+ #Return(dataall)\n+\t\n+\t\t}\n+\t}\n+\telse{\n+\t\tprint("The files are not the same length")\n+\t}\n+}\n+\n+args <- commandArgs(trailingOnly = TRUE)\n+print(args)\n+\t\t\n+RNAseqDataAnnotation(args[1], args[2],args[3], args[4], args[5], args[6], args[7])\n+\n+#R --slave --vanilla --verbose --file=/home/lornage/Bureau/Pour_galaxy/RNAseqDataAnnotation.R --args /home/lornage/Bureau/Test_function /home/lornage/Bureau/ichierconvertitnames.txt Homo_sapiens 75 /home/lornage/Bureau/testttttt5.txt /home/lornage/Bureau/Script_R/Ensembl_Version_Host.txt /home/lornage/Bureau/Script_R/Ensemble_Specie_Dataset.txt\n+\n+\n+\n+\n+\n+\n' |
| b |
| diff -r 80216215973c -r 73ea91916c9d RNAseqDataAnnotation/RNAseqDataAnnotation.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/RNAseqDataAnnotation/RNAseqDataAnnotation.xml Wed Nov 19 12:10:09 2014 -0500 |
| b |
| @@ -0,0 +1,52 @@ +<tool id="RNAseqDataAnnotation" name="RNAseqDataAnnotation" version="1.0.0"> + <description>tool for RNAseq Data Normalisation and Annotation</description> + <requirements> + <!--<requirement type="set_environment">SCRIPT_PATH</requirement>--> + <requirement type="package" version="3.0.2">R_3_0_2</requirement> + <requirement type="package" version="1.0">DESeq2biomaRt</requirement> + </requirements> + + <command> + R --slave --vanilla --file=RNAseqDataAnnotation.R --args + $path2htseqfiles + $samplenamefile + $Species + $ensversion + $conversionensemblversion + $conversionensemblname + $fileout + </command> + + <inputs> + <param name="path2htseqfiles" label="Path to the directory containing the files from HTSeq-count" type="text"/> + <param name="samplenamefile" label="Conversion file sample/conditions" type="data" format="tabular" help="file should be tab-delimited"/> + <param name="Species" type="select" label="Select the specie for your data" help="If your specie of interest is not listed, your data will be normalized but no annotation will be added. Contact us if you want us to add your specie." > + <option value="Homo_sapiens">Homo sapiens</option> + <option value="Mus_musculus">Mus musculus</option> + <option value="">Other specie</option> + </param> + <param name="ensversion" type="select" label="Select the version of Ensembl to use" > + <option value="67">Version 67</option> + <option value="68">Version 68</option> + <option value="69">Version 69</option> + <option value="70">Version 70</option> + <option value="71">Version 71</option> + <option value="72">Version 72</option> + <option value="73">Version 73</option> + <option value="74">Version 74</option> + <option value="75">Version 75</option> + <option value="76">Version 76</option> + <option value="77">Version 77</option> + </param> + <param name="conversionensemblversion" label="File for conversion Ensembl to version" type="data" format="tabular" help="Tab-delimited input file" /> + <param name="conversionensemblname" label="File for conversion Ensemble name of the specie " type="data" format="tabular" help="Tab-delimited input file"/> + </inputs> + + <outputs> +<param name="fileout" label="Path where the resulting file should be stored" type="data" format="tabular"/> + </outputs> + <help> +**What it does* +**Example** + </help> + </tool> |
| b |
| diff -r 80216215973c -r 73ea91916c9d RNAseqDataAnnotation/packages.R --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/RNAseqDataAnnotation/packages.R Wed Nov 19 12:10:09 2014 -0500 |
| b |
| @@ -0,0 +1,5 @@ +source("http://bioconductor.org/biocLite.R") +biocLite() +biocLite("DESeq2") +biocLite("biomaRt") +install.packages("plyr") |
| b |
| diff -r 80216215973c -r 73ea91916c9d RNAseqDataAnnotation/tool_dependencies.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/RNAseqDataAnnotation/tool_dependencies.xml Wed Nov 19 12:10:09 2014 -0500 |
| b |
| @@ -0,0 +1,22 @@ +<?xml version="1.0"?> +<tool_dependency> + <package name="R_3_0_2" version="3.0.2"> + <repository changeset_revision="b6fe8ca3230d" name="package_r_3_0_2" owner="iuc" prior_installation_required="True" toolshed="https://testtoolshed.g2.bx.psu.edu" /> + </package> + <package name="DESeq2biomaRt" version="1.0"> + <install version="1.0"> + <actions> + <action type="set_environment_for_install"> + <repository changeset_revision="b6fe8ca3230d" name="package_r_3_0_2" owner="iuc" toolshed="https://testtoolshed.g2.bx.psu.edu"> + <package name="R_3_0_2" version="3.0.2" /> + </repository> + </action> + <action type="shell_command">R CMD BATCH packages.R</action> + <action type="shell_command">echo "export PATH=$PATH" > $INSTALL_DIR/env.sh </action> + <action type="shell_command">chmod 755 $INSTALL_DIR/env.sh </action> + </actions> + </install> + <readme> + </readme> + </package> +</tool_dependency> |