Mercurial > repos > davidvanzessen > argalaxy_tools

changeset 40:106275b54470 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Tue, 30 May 2017 07:26:33 -0400
parents bad6a9a53ae7
children a70dbcfa5d8a
files experimental_design/experimental_design.r imgt_loader/imgt_loader.r imgt_loader/imgt_loader.sh report_clonality/RScript.r report_clonality/r_wrapper.sh
diffstat 5 files changed, 50 insertions(+), 13 deletions(-) [+]
line wrap: on
line diff
--- a/experimental_design/experimental_design.r	Mon Apr 24 09:39:59 2017 -0400
+++ b/experimental_design/experimental_design.r	Tue May 30 07:26:33 2017 -0400
@@ -15,6 +15,11 @@
 		print(paste("Adding file", counter, "to", current.id))
 		dat = read.table(current, sep="\t", header=T, quote="", fill=T)
 		
+		if(nrow(dat) == 0){
+			print(paste(counter, "of", current.id, "has no sequences, skipping"))
+			next
+		}
+		
 		#IMGT check
 		
 		dat$Sample = current.id
--- a/imgt_loader/imgt_loader.r	Mon Apr 24 09:39:59 2017 -0400
+++ b/imgt_loader/imgt_loader.r	Tue May 30 07:26:33 2017 -0400
@@ -1,4 +1,5 @@
 args <- commandArgs(trailingOnly = TRUE)
+options(show.error.locations = TRUE)
 
 summ.file = args[1]
 sequences.file = args[2]
@@ -7,12 +8,19 @@
 gapped.aa.file = args[5]
 out.file = args[6]
 
+print(summ.file)
+print(out.file)
+
 summ = read.table(summ.file, sep="\t", header=T, quote="", fill=T)
 sequences = read.table(sequences.file, sep="\t", header=T, quote="", fill=T)
 aa = read.table(aa.file, sep="\t", header=T, quote="", fill=T)
 gapped.aa = read.table(gapped.aa.file, sep="\t", header=T, quote="", fill=T)
 junction = read.table(junction.file, sep="\t", header=T, quote="", fill=T)
 
+print(paste("nrow(summ)", nrow(summ)))
+
+write.table(summ, out.file, sep="\t", quote=F, row.names=F, col.names=T)
+
 fix_column_names = function(df){
     if("V.DOMAIN.Functionality" %in% names(df)){
         names(df)[names(df) == "V.DOMAIN.Functionality"] = "Functionality"
@@ -31,6 +39,8 @@
 gapped.aa = fix_column_names(gapped.aa)
 junction = fix_column_names(junction)
 
+print(paste("nrow(summ)", nrow(summ)))
+
 old_summary_columns=c('Sequence.ID','JUNCTION.frame','V.GENE.and.allele','D.GENE.and.allele','J.GENE.and.allele','CDR1.IMGT.length','CDR2.IMGT.length','CDR3.IMGT.length','Orientation')
 old_sequence_columns=c('CDR1.IMGT','CDR2.IMGT','CDR3.IMGT')
 old_junction_columns=c('JUNCTION')
@@ -43,6 +53,9 @@
 
 out=summ[,c("Sequence.ID","JUNCTION.frame","V.GENE.and.allele","D.GENE.and.allele","J.GENE.and.allele")]
 
+print(paste("nrow(summ)", nrow(summ)))
+print(paste("nrow(aa)", nrow(aa)))
+
 out[,"CDR1.Seq"] = aa[,"CDR1.IMGT"]
 out[,"CDR1.Length"] = summ[,"CDR1.IMGT.length"]
 
@@ -78,24 +91,20 @@
 names(out) = c('ID','VDJ Frame','Top V Gene','Top D Gene','Top J Gene','CDR1 Seq','CDR1 Length','CDR2 Seq','CDR2 Length','CDR3 Seq','CDR3 Length','CDR3 Seq DNA','CDR3 Length DNA','Strand','CDR3 Found How','Functionality','V-REGION identity %','V-REGION identity nt','D-REGION reading frame','AA JUNCTION','Functionality comment','Sequence','FR1-IMGT','FR2-IMGT','FR3-IMGT','CDR3-IMGT','JUNCTION','J-REGION','FR4-IMGT','P3V-nt nb','N-REGION-nt nb','N1-REGION-nt nb','P5D-nt nb','P3D-nt nb','N2-REGION-nt nb','P5J-nt nb','3V-REGION trimmed-nt nb','5D-REGION trimmed-nt nb','3D-REGION trimmed-nt nb','5J-REGION trimmed-nt nb','N-REGION','N1-REGION','N2-REGION', 'P5.D1.nt.nb', 'P3.D1.nt.nb', 'N2.REGION.nt.nb', 'P5.D2.nt.nb', 'P3.D2.nt.nb', 'N3.REGION.nt.nb', 'P5.D3.nt.nb', 'P3.D2.nt.nb', 'N4.REGION.nt.nb', 'X5.D1.REGION.trimmed.nt.nb', 'X3.D1.REGION.trimmed.nt.nb', 'X5.D2.REGION.trimmed.nt.nb', 'X3.D2.REGION.trimmed.nt.nb', 'X5.D3.REGION.trimmed.nt.nb', 'X3.D3.REGION.trimmed.nt.nb', 'D.REGION.nt.nb', 'D1.REGION.nt.nb', 'D2.REGION.nt.nb', 'D3.REGION.nt.nb')
 
 out[,"VDJ Frame"] = as.character(out[,"VDJ Frame"])
-
 fltr = out[,"VDJ Frame"] == "in-frame"
-if(any(fltr)){
+if(any(fltr, na.rm=T)){
 	out[fltr, "VDJ Frame"] = "In-frame"
 }
-
 fltr = out[,"VDJ Frame"] == "null"
-if(any(fltr)){
+if(any(fltr, na.rm = T)){
 	out[fltr, "VDJ Frame"] = "Out-of-frame"
 }
-
 fltr = out[,"VDJ Frame"] == "out-of-frame"
-if(any(fltr)){
+if(any(fltr, na.rm = T)){
 	out[fltr, "VDJ Frame"] = "Out-of-frame"
 }
-
 fltr = out[,"VDJ Frame"] == ""
-if(any(fltr)){
+if(any(fltr, na.rm = T)){
 	out[fltr, "VDJ Frame"] = "Out-of-frame"
 }
 
@@ -103,7 +112,7 @@
 	out[,col] = as.character(out[,col])
 	fltr = out[,col] == ""
 	fltr[is.na(fltr)] = T
-	if(any(fltr)){
+	if(any(fltr, na.rm = T)){
 		out[fltr,col] = "NA"
 	}
 }
--- a/imgt_loader/imgt_loader.sh	Mon Apr 24 09:39:59 2017 -0400
+++ b/imgt_loader/imgt_loader.sh	Tue May 30 07:26:33 2017 -0400
@@ -66,6 +66,9 @@
 find $PWD/$name/files -iname "5_*" -exec cat {} + > $PWD/$name/aa.txt
 find $PWD/$name/files -iname "6_*" -exec cat {} + > $PWD/$name/junction.txt
 
+echo "summ.txt `cat $PWD/$name/summ.txt | wc -l`"
+echo "aa.txt `cat $PWD/$name/aa.txt | wc -l`"
+
 #python $dir/imgt_loader.py --summ $PWD/$name/summ.txt --aa $PWD/$name/aa.txt --junction $PWD/$name/junction.txt --output $output
 
 Rscript --verbose $dir/imgt_loader.r $PWD/$name/summ.txt $PWD/$name/sequences.txt $PWD/$name/aa.txt $PWD/$name/junction.txt $PWD/$name/gapped_aa.txt $output 2>&1
--- a/report_clonality/RScript.r	Mon Apr 24 09:39:59 2017 -0400
+++ b/report_clonality/RScript.r	Tue May 30 07:26:33 2017 -0400
@@ -122,12 +122,17 @@
   clonalityFrame = clonalityFrame[!duplicated(clonalityFrame$clonality_clonaltype), ]
 }
 
+if(nrow(PRODF) == 0){
+	stop("No sequences left after filtering")
+}
+
 prod.unique.sample.count = data.frame(data.table(PRODF)[, list(Productive_unique=.N), by=c("Sample")])
 prod.unique.rep.count = data.frame(data.table(PRODF)[, list(Productive_unique=.N), by=c("Sample", "Replicate")])
 
 unprod.unique.sample.count = data.frame(data.table(UNPROD)[, list(Unproductive_unique=.N), by=c("Sample")])
 unprod.unique.rep.count = data.frame(data.table(UNPROD)[, list(Unproductive_unique=.N), by=c("Sample", "Replicate")])
 
+
 PRODF$freq = 1
 
 if(any(grepl(pattern="_", x=PRODF$ID))){ #the frequency can be stored in the ID with the pattern ".*_freq_.*"
@@ -176,6 +181,7 @@
 
 #create the table on the overview page with the productive/unique counts per sample/replicate
 #first for sample
+
 sample.count = merge(input.sample.count, prod.sample.count, by="Sample", all.x=T)
 sample.count$perc_prod = round(sample.count$Productive / sample.count$All * 100)
 sample.count = merge(sample.count, prod.unique.sample.count, by="Sample", all.x=T)
@@ -188,6 +194,16 @@
 
 #then sample/replicate
 rep.count = merge(input.rep.count, prod.rep.count, by=c("Sample", "Replicate"), all.x=T)
+
+print(rep.count)
+
+fltr = is.na(rep.count$Productive)
+if(any(fltr)){
+	rep.count[fltr,"Productive"] = 0
+}
+
+print(rep.count)
+
 rep.count$perc_prod = round(rep.count$Productive / rep.count$All * 100)
 rep.count = merge(rep.count, prod.unique.rep.count, by=c("Sample", "Replicate"), all.x=T)
 rep.count$perc_prod_un = round(rep.count$Productive_unique / rep.count$All * 100)
--- a/report_clonality/r_wrapper.sh	Mon Apr 24 09:39:59 2017 -0400
+++ b/report_clonality/r_wrapper.sh	Tue May 30 07:26:33 2017 -0400
@@ -42,10 +42,14 @@
 	do
 		echo "<tr><td>$sample</td>" >> $2
 		echo "<td>$all</td>" >> $2
-		echo "<td>$productive (${perc_prod}%)</td>" >> $2
-		echo "<td>$productive_unique (${perc_prod_un}%)</td>" >> $2
-		echo "<td>$unproductive (${perc_unprod}%)</td>" >> $2
-		echo "<td>$unproductive_unique (${perc_unprod_un}%)</td></tr>" >> $2
+		if [[ "$productive" != "0" ]] ; then
+			echo "<td>$productive (${perc_prod}%)</td>" >> $2
+			echo "<td>$productive_unique (${perc_prod_un}%)</td>" >> $2
+			echo "<td>$unproductive (${perc_unprod}%)</td>" >> $2
+			echo "<td>$unproductive_unique (${perc_unprod_un}%)</td></tr>" >> $2
+		else
+			echo "<td colspan='4' style='background-color: red;'>No productive sequences!</td>" >> $2
+		fi
 done < $outputDir/productive_counting.txt
 echo "</table><br />" >> $2
 echo "Table showing the number and percentage of (unique) productive and unproductive sequences per donor and per replicate. <br />" >> $2