comparison rgedgeR/rgedgeRpaired.xml @ 13:f542654faa3b draft

Uploaded

12:aa528c834664

  nzN = nrow(nonzerod)
  nzrs = rowSums(nonzerod)
  zN = allN - nzN
  print('# Quantiles for non-zero row counts:',quote=F)
  print(quantile(nzrs,probs=seq(0,1,0.1)),quote=F)
  if (useNDF == "T")
  {
    gt1rpin3 = rowSums(Count_Matrix/expandAsMatrix(colTotmillionreads,dim(Count_Matrix)) >= 1) >= nscut
    lo = colSums(Count_Matrix[!gt1rpin3,])
    workCM = Count_Matrix[gt1rpin3,]
    cleanrs = rowSums(workCM)

  dev.off()
  colnames(normData) = paste( colnames(normData),'N',sep="_")
  print(paste('Raw sample read totals',paste(colSums(nonzerod,na.rm=T),collapse=',')))
  nzd = data.frame(log(nonzerod + 1e-2,10))
  boxPlot(rawrs=nzd,cleanrs=log(normData,10),maint='TMM Normalisation',myTitle=myTitle,pdfname="edgeR_raw_norm_counts_box.pdf")
  write.table(soutput,out_edgeR, quote=FALSE, sep="\t",row.names=F)
  tt = cbind( 
    Name=as.character(rownames(DGEList\$counts)),
    DE\$table,
    adj.p.value=p.adjust(DE\$table\$PValue, method=fdrtype),
    Dispersion=DGEList\$tagwise.dispersion,totreads=cmrowsums

    rDESeq = cbind(Contig=rownames(workCM),rDESeq,NReads=cmrowsums,URL=contigurls)
    srDESeq = rDESeq[order(rDESeq\$pvalue),]
    qqPlot(descr=paste(myTitle,'DESeq2 qqplot'),pvector=rDESeq\$pvalue,outpdf='DESeq2_qqplot.pdf')
    cat("# DESeq top 50\n")
    print.noquote(srDESeq[1:50,])
    write.table(srDESeq,out_DESeq2, quote=FALSE, sep="\t",row.names=F)
    topresults.DESeq = rDESeq[which(rDESeq\$padj < fdrthresh), ]
    DESeqcountsindex = which(allgenes %in% rownames(topresults.DESeq))
    DESeqcounts = rep(0, length(allgenes))
    DESeqcounts[DESeqcountsindex] = 1
    pdf("DESeq2_dispersion_estimates.pdf")

      rvoom = topTable(fit, coef = length(colnames(mydesign)), adj = fdrtype, n = Inf, sort="none")
      qqPlot(descr=paste(myTitle,'VOOM-limma QQ plot'),pvector=rvoom\$P.Value,outpdf='VOOM_qqplot.pdf')
      rownames(rvoom) = rownames(workCM)
      rvoom = cbind(rvoom,NReads=cmrowsums,URL=contigurls)
      srvoom = rvoom[order(rvoom\$P.Value),]
      write.table(srvoom,out_VOOM, quote=FALSE, sep="\t",row.names=F)
      # Use an FDR cutoff to find interesting samples for edgeR, DESeq and voom/limma
      topresults.voom = srvoom[which(rvoom\$adj.P.Val < fdrthresh), ]
      voomcountsindex = which(allgenes %in% topresults.voom\$ID)
      voomcounts = rep(0, length(allgenes))
      voomcounts[voomcountsindex] = 1
      cat("# VOOM top 50\n")
      print(srvoom[1:50,])
      sink()
  }

  if (doCamera) {
  doGSEA(y=DGEList,design=mydesign,histgmt=histgmt,bigmt=bigmt,ntest=20,myTitle=myTitle,

###sink(stdout(),append=T,type="message")
builtin_gmt=""
history_gmt=""
out_edgeR = F
out_DESeq2 = F
out_VOOM = F
doDESeq2 = $DESeq2.doDESeq2 # make these T or F
doVoom = $doVoom
doCamera = F 
doedgeR = $edgeR.doedgeR
edgeR_priordf = 0

#if $DESeq2.doDESeq2 == "T":
  out_DESeq2 = "$out_DESeq2"
  DESeq_fitType = "$DESeq2.DESeq_fitType"
#end if
#if $edgeR.doedgeR == "T":
  out_edgeR = "$out_edgeR"
  edgeR_priordf = $edgeR.edgeR_priordf
#end if
#if $doVoom == "T":
  out_VOOM = "$out_VOOM"
#end if

Out_Dir = "$html_file.files_path"
Input =  "$input1"
TreatmentName = "$treatment_name"

ControlCols= "$Control_cols"
org = "$input1.dbkey"
if (org == "") { org = "hg19"}
fdrtype = "$fdrtype"
fdrthresh = $fdrthresh
useNDF = "$useNDF"
fQ = $fQ # non-differential centile cutoff
myTitle = "$title"
subjects = c($subjectids)
nsubj = length(subjects)
if (nsubj > 0) {

LibSizes = Count_Matrix[subset(rn,islib),][1] # take first
Count_Matrix = Count_Matrix[subset(rn,! islib),]
group = c(rep(TreatmentName,length(TCols)), rep(ControlName,length(CCols)) )             #Build a group descriptor
group = factor(group, levels=c(ControlName,TreatmentName))
colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_")                   #Relable columns
results = edgeIt(Count_Matrix=Count_Matrix,group=group,out_edgeR=out_edgeR, out_VOOM=out_VOOM, out_DESeq2=out_DESeq2,
                 fdrtype='BH',priordf=edgeR_priordf,fdrthresh=0.05,outputdir='.',
                 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=c(),
                 doDESeq2=doDESeq2,doVoom=doVoom,doCamera=doCamera,doedgeR=doedgeR,org=org,
                 histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType)
sessionInfo()

13:f542654faa3b

  nzN = nrow(nonzerod)
  nzrs = rowSums(nonzerod)
  zN = allN - nzN
  print('# Quantiles for non-zero row counts:',quote=F)
  print(quantile(nzrs,probs=seq(0,1,0.1)),quote=F)
  if (useNDF == T)
  {
    gt1rpin3 = rowSums(Count_Matrix/expandAsMatrix(colTotmillionreads,dim(Count_Matrix)) >= 1) >= nscut
    lo = colSums(Count_Matrix[!gt1rpin3,])
    workCM = Count_Matrix[gt1rpin3,]
    cleanrs = rowSums(workCM)

  dev.off()
  colnames(normData) = paste( colnames(normData),'N',sep="_")
  print(paste('Raw sample read totals',paste(colSums(nonzerod,na.rm=T),collapse=',')))
  nzd = data.frame(log(nonzerod + 1e-2,10))
  boxPlot(rawrs=nzd,cleanrs=log(normData,10),maint='TMM Normalisation',myTitle=myTitle,pdfname="edgeR_raw_norm_counts_box.pdf")
  write.table(soutput,file=out_edgeR, quote=FALSE, sep="\t",row.names=F)
  tt = cbind( 
    Name=as.character(rownames(DGEList\$counts)),
    DE\$table,
    adj.p.value=p.adjust(DE\$table\$PValue, method=fdrtype),
    Dispersion=DGEList\$tagwise.dispersion,totreads=cmrowsums

    rDESeq = cbind(Contig=rownames(workCM),rDESeq,NReads=cmrowsums,URL=contigurls)
    srDESeq = rDESeq[order(rDESeq\$pvalue),]
    qqPlot(descr=paste(myTitle,'DESeq2 qqplot'),pvector=rDESeq\$pvalue,outpdf='DESeq2_qqplot.pdf')
    cat("# DESeq top 50\n")
    print.noquote(srDESeq[1:50,])
    write.table(srDESeq,file=out_DESeq2, quote=FALSE, sep="\t",row.names=F)
    topresults.DESeq = rDESeq[which(rDESeq\$padj < fdrthresh), ]
    DESeqcountsindex = which(allgenes %in% rownames(topresults.DESeq))
    DESeqcounts = rep(0, length(allgenes))
    DESeqcounts[DESeqcountsindex] = 1
    pdf("DESeq2_dispersion_estimates.pdf")

      rvoom = topTable(fit, coef = length(colnames(mydesign)), adj = fdrtype, n = Inf, sort="none")
      qqPlot(descr=paste(myTitle,'VOOM-limma QQ plot'),pvector=rvoom\$P.Value,outpdf='VOOM_qqplot.pdf')
      rownames(rvoom) = rownames(workCM)
      rvoom = cbind(rvoom,NReads=cmrowsums,URL=contigurls)
      srvoom = rvoom[order(rvoom\$P.Value),]
      cat("# VOOM top 50\n")
      print(srvoom[1:50,])
      write.table(srvoom,file=out_VOOM, quote=FALSE, sep="\t",row.names=F)
      # Use an FDR cutoff to find interesting samples for edgeR, DESeq and voom/limma
      topresults.voom = srvoom[which(rvoom\$adj.P.Val < fdrthresh), ]
      voomcountsindex = which(allgenes %in% topresults.voom\$ID)
      voomcounts = rep(0, length(allgenes))
      voomcounts[voomcountsindex] = 1
      sink()
  }

  if (doCamera) {
  doGSEA(y=DGEList,design=mydesign,histgmt=histgmt,bigmt=bigmt,ntest=20,myTitle=myTitle,

###sink(stdout(),append=T,type="message")
builtin_gmt=""
history_gmt=""
out_edgeR = F
out_DESeq2 = F
out_VOOM = "${out_VOOM}"
doDESeq2 = $DESeq2.doDESeq2 # make these T or F
doVoom = $doVoom
doCamera = F 
doedgeR = $edgeR.doedgeR
edgeR_priordf = 0


#if ${doVoom} == "T":
  out_VOOM = "${out_VOOM}"
#end if

#if ${DESeq2.doDESeq2} == "T":
  out_DESeq2 = "${out_DESeq2}"
  DESeq_fitType = "${DESeq2.DESeq_fitType}"
#end if

#if ${edgeR.doedgeR} == "T":
  out_edgeR = "${out_edgeR}"
  edgeR_priordf = ${edgeR.edgeR_priordf}
#end if

Out_Dir = "$html_file.files_path"
Input =  "$input1"
TreatmentName = "$treatment_name"

ControlCols= "$Control_cols"
org = "$input1.dbkey"
if (org == "") { org = "hg19"}
fdrtype = "$fdrtype"
fdrthresh = $fdrthresh
useNDF = $useNDF
fQ = $fQ # non-differential centile cutoff
myTitle = "$title"
subjects = c($subjectids)
nsubj = length(subjects)
if (nsubj > 0) {

LibSizes = Count_Matrix[subset(rn,islib),][1] # take first
Count_Matrix = Count_Matrix[subset(rn,! islib),]
group = c(rep(TreatmentName,length(TCols)), rep(ControlName,length(CCols)) )             #Build a group descriptor
group = factor(group, levels=c(ControlName,TreatmentName))
colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_")                   #Relable columns
results = edgeIt(Count_Matrix=Count_Matrix,group=group, out_edgeR=out_edgeR, out_VOOM=out_VOOM, out_DESeq2=out_DESeq2,
                 fdrtype='BH',priordf=edgeR_priordf,fdrthresh=0.05,outputdir='.',
                 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=c(),
                 doDESeq2=doDESeq2,doVoom=doVoom,doCamera=doCamera,doedgeR=doedgeR,org=org,
                 histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType)
sessionInfo()