comparison rgedgeR/rgedgeRpaired.xml @ 13:f542654faa3b draft

Uploaded
author fubar
date Sat, 27 Jul 2013 02:18:41 -0400
parents aa528c834664
children 8272815ba8ce
comparison
equal deleted inserted replaced
12:aa528c834664 13:f542654faa3b
399 nzN = nrow(nonzerod) 399 nzN = nrow(nonzerod)
400 nzrs = rowSums(nonzerod) 400 nzrs = rowSums(nonzerod)
401 zN = allN - nzN 401 zN = allN - nzN
402 print('# Quantiles for non-zero row counts:',quote=F) 402 print('# Quantiles for non-zero row counts:',quote=F)
403 print(quantile(nzrs,probs=seq(0,1,0.1)),quote=F) 403 print(quantile(nzrs,probs=seq(0,1,0.1)),quote=F)
404 if (useNDF == "T") 404 if (useNDF == T)
405 { 405 {
406 gt1rpin3 = rowSums(Count_Matrix/expandAsMatrix(colTotmillionreads,dim(Count_Matrix)) >= 1) >= nscut 406 gt1rpin3 = rowSums(Count_Matrix/expandAsMatrix(colTotmillionreads,dim(Count_Matrix)) >= 1) >= nscut
407 lo = colSums(Count_Matrix[!gt1rpin3,]) 407 lo = colSums(Count_Matrix[!gt1rpin3,])
408 workCM = Count_Matrix[gt1rpin3,] 408 workCM = Count_Matrix[gt1rpin3,]
409 cleanrs = rowSums(workCM) 409 cleanrs = rowSums(workCM)
510 dev.off() 510 dev.off()
511 colnames(normData) = paste( colnames(normData),'N',sep="_") 511 colnames(normData) = paste( colnames(normData),'N',sep="_")
512 print(paste('Raw sample read totals',paste(colSums(nonzerod,na.rm=T),collapse=','))) 512 print(paste('Raw sample read totals',paste(colSums(nonzerod,na.rm=T),collapse=',')))
513 nzd = data.frame(log(nonzerod + 1e-2,10)) 513 nzd = data.frame(log(nonzerod + 1e-2,10))
514 boxPlot(rawrs=nzd,cleanrs=log(normData,10),maint='TMM Normalisation',myTitle=myTitle,pdfname="edgeR_raw_norm_counts_box.pdf") 514 boxPlot(rawrs=nzd,cleanrs=log(normData,10),maint='TMM Normalisation',myTitle=myTitle,pdfname="edgeR_raw_norm_counts_box.pdf")
515 write.table(soutput,out_edgeR, quote=FALSE, sep="\t",row.names=F) 515 write.table(soutput,file=out_edgeR, quote=FALSE, sep="\t",row.names=F)
516 tt = cbind( 516 tt = cbind(
517 Name=as.character(rownames(DGEList\$counts)), 517 Name=as.character(rownames(DGEList\$counts)),
518 DE\$table, 518 DE\$table,
519 adj.p.value=p.adjust(DE\$table\$PValue, method=fdrtype), 519 adj.p.value=p.adjust(DE\$table\$PValue, method=fdrtype),
520 Dispersion=DGEList\$tagwise.dispersion,totreads=cmrowsums 520 Dispersion=DGEList\$tagwise.dispersion,totreads=cmrowsums
565 rDESeq = cbind(Contig=rownames(workCM),rDESeq,NReads=cmrowsums,URL=contigurls) 565 rDESeq = cbind(Contig=rownames(workCM),rDESeq,NReads=cmrowsums,URL=contigurls)
566 srDESeq = rDESeq[order(rDESeq\$pvalue),] 566 srDESeq = rDESeq[order(rDESeq\$pvalue),]
567 qqPlot(descr=paste(myTitle,'DESeq2 qqplot'),pvector=rDESeq\$pvalue,outpdf='DESeq2_qqplot.pdf') 567 qqPlot(descr=paste(myTitle,'DESeq2 qqplot'),pvector=rDESeq\$pvalue,outpdf='DESeq2_qqplot.pdf')
568 cat("# DESeq top 50\n") 568 cat("# DESeq top 50\n")
569 print.noquote(srDESeq[1:50,]) 569 print.noquote(srDESeq[1:50,])
570 write.table(srDESeq,out_DESeq2, quote=FALSE, sep="\t",row.names=F) 570 write.table(srDESeq,file=out_DESeq2, quote=FALSE, sep="\t",row.names=F)
571 topresults.DESeq = rDESeq[which(rDESeq\$padj < fdrthresh), ] 571 topresults.DESeq = rDESeq[which(rDESeq\$padj < fdrthresh), ]
572 DESeqcountsindex = which(allgenes %in% rownames(topresults.DESeq)) 572 DESeqcountsindex = which(allgenes %in% rownames(topresults.DESeq))
573 DESeqcounts = rep(0, length(allgenes)) 573 DESeqcounts = rep(0, length(allgenes))
574 DESeqcounts[DESeqcountsindex] = 1 574 DESeqcounts[DESeqcountsindex] = 1
575 pdf("DESeq2_dispersion_estimates.pdf") 575 pdf("DESeq2_dispersion_estimates.pdf")
621 rvoom = topTable(fit, coef = length(colnames(mydesign)), adj = fdrtype, n = Inf, sort="none") 621 rvoom = topTable(fit, coef = length(colnames(mydesign)), adj = fdrtype, n = Inf, sort="none")
622 qqPlot(descr=paste(myTitle,'VOOM-limma QQ plot'),pvector=rvoom\$P.Value,outpdf='VOOM_qqplot.pdf') 622 qqPlot(descr=paste(myTitle,'VOOM-limma QQ plot'),pvector=rvoom\$P.Value,outpdf='VOOM_qqplot.pdf')
623 rownames(rvoom) = rownames(workCM) 623 rownames(rvoom) = rownames(workCM)
624 rvoom = cbind(rvoom,NReads=cmrowsums,URL=contigurls) 624 rvoom = cbind(rvoom,NReads=cmrowsums,URL=contigurls)
625 srvoom = rvoom[order(rvoom\$P.Value),] 625 srvoom = rvoom[order(rvoom\$P.Value),]
626 write.table(srvoom,out_VOOM, quote=FALSE, sep="\t",row.names=F) 626 cat("# VOOM top 50\n")
627 print(srvoom[1:50,])
628 write.table(srvoom,file=out_VOOM, quote=FALSE, sep="\t",row.names=F)
627 # Use an FDR cutoff to find interesting samples for edgeR, DESeq and voom/limma 629 # Use an FDR cutoff to find interesting samples for edgeR, DESeq and voom/limma
628 topresults.voom = srvoom[which(rvoom\$adj.P.Val < fdrthresh), ] 630 topresults.voom = srvoom[which(rvoom\$adj.P.Val < fdrthresh), ]
629 voomcountsindex = which(allgenes %in% topresults.voom\$ID) 631 voomcountsindex = which(allgenes %in% topresults.voom\$ID)
630 voomcounts = rep(0, length(allgenes)) 632 voomcounts = rep(0, length(allgenes))
631 voomcounts[voomcountsindex] = 1 633 voomcounts[voomcountsindex] = 1
632 cat("# VOOM top 50\n")
633 print(srvoom[1:50,])
634 sink() 634 sink()
635 } 635 }
636 636
637 if (doCamera) { 637 if (doCamera) {
638 doGSEA(y=DGEList,design=mydesign,histgmt=histgmt,bigmt=bigmt,ntest=20,myTitle=myTitle, 638 doGSEA(y=DGEList,design=mydesign,histgmt=histgmt,bigmt=bigmt,ntest=20,myTitle=myTitle,
669 ###sink(stdout(),append=T,type="message") 669 ###sink(stdout(),append=T,type="message")
670 builtin_gmt="" 670 builtin_gmt=""
671 history_gmt="" 671 history_gmt=""
672 out_edgeR = F 672 out_edgeR = F
673 out_DESeq2 = F 673 out_DESeq2 = F
674 out_VOOM = F 674 out_VOOM = "${out_VOOM}"
675 doDESeq2 = $DESeq2.doDESeq2 # make these T or F 675 doDESeq2 = $DESeq2.doDESeq2 # make these T or F
676 doVoom = $doVoom 676 doVoom = $doVoom
677 doCamera = F 677 doCamera = F
678 doedgeR = $edgeR.doedgeR 678 doedgeR = $edgeR.doedgeR
679 edgeR_priordf = 0 679 edgeR_priordf = 0
680 680
681 #if $DESeq2.doDESeq2 == "T": 681
682 out_DESeq2 = "$out_DESeq2" 682 #if ${doVoom} == "T":
683 DESeq_fitType = "$DESeq2.DESeq_fitType" 683 out_VOOM = "${out_VOOM}"
684 #end if 684 #end if
685 #if $edgeR.doedgeR == "T": 685
686 out_edgeR = "$out_edgeR" 686 #if ${DESeq2.doDESeq2} == "T":
687 edgeR_priordf = $edgeR.edgeR_priordf 687 out_DESeq2 = "${out_DESeq2}"
688 DESeq_fitType = "${DESeq2.DESeq_fitType}"
688 #end if 689 #end if
689 #if $doVoom == "T": 690
690 out_VOOM = "$out_VOOM" 691 #if ${edgeR.doedgeR} == "T":
692 out_edgeR = "${out_edgeR}"
693 edgeR_priordf = ${edgeR.edgeR_priordf}
691 #end if 694 #end if
692 695
693 Out_Dir = "$html_file.files_path" 696 Out_Dir = "$html_file.files_path"
694 Input = "$input1" 697 Input = "$input1"
695 TreatmentName = "$treatment_name" 698 TreatmentName = "$treatment_name"
698 ControlCols= "$Control_cols" 701 ControlCols= "$Control_cols"
699 org = "$input1.dbkey" 702 org = "$input1.dbkey"
700 if (org == "") { org = "hg19"} 703 if (org == "") { org = "hg19"}
701 fdrtype = "$fdrtype" 704 fdrtype = "$fdrtype"
702 fdrthresh = $fdrthresh 705 fdrthresh = $fdrthresh
703 useNDF = "$useNDF" 706 useNDF = $useNDF
704 fQ = $fQ # non-differential centile cutoff 707 fQ = $fQ # non-differential centile cutoff
705 myTitle = "$title" 708 myTitle = "$title"
706 subjects = c($subjectids) 709 subjects = c($subjectids)
707 nsubj = length(subjects) 710 nsubj = length(subjects)
708 if (nsubj > 0) { 711 if (nsubj > 0) {
738 LibSizes = Count_Matrix[subset(rn,islib),][1] # take first 741 LibSizes = Count_Matrix[subset(rn,islib),][1] # take first
739 Count_Matrix = Count_Matrix[subset(rn,! islib),] 742 Count_Matrix = Count_Matrix[subset(rn,! islib),]
740 group = c(rep(TreatmentName,length(TCols)), rep(ControlName,length(CCols)) ) #Build a group descriptor 743 group = c(rep(TreatmentName,length(TCols)), rep(ControlName,length(CCols)) ) #Build a group descriptor
741 group = factor(group, levels=c(ControlName,TreatmentName)) 744 group = factor(group, levels=c(ControlName,TreatmentName))
742 colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_") #Relable columns 745 colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_") #Relable columns
743 results = edgeIt(Count_Matrix=Count_Matrix,group=group,out_edgeR=out_edgeR, out_VOOM=out_VOOM, out_DESeq2=out_DESeq2, 746 results = edgeIt(Count_Matrix=Count_Matrix,group=group, out_edgeR=out_edgeR, out_VOOM=out_VOOM, out_DESeq2=out_DESeq2,
744 fdrtype='BH',priordf=edgeR_priordf,fdrthresh=0.05,outputdir='.', 747 fdrtype='BH',priordf=edgeR_priordf,fdrthresh=0.05,outputdir='.',
745 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=c(), 748 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=c(),
746 doDESeq2=doDESeq2,doVoom=doVoom,doCamera=doCamera,doedgeR=doedgeR,org=org, 749 doDESeq2=doDESeq2,doVoom=doVoom,doCamera=doCamera,doedgeR=doedgeR,org=org,
747 histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType) 750 histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType)
748 sessionInfo() 751 sessionInfo()