Merge pull request #485 from szilvajuhos/master

Fixing #357 also better speedseq filter for FreeBayes

Author: maxulysse

main.nf

@@ -1307,6 +1307,8 @@ process RunAscat {
 
   script:
   """
+  # get rid of "chr" string if there is any
+  for f in *BAF *LogR; do sed 's/chr//g' \$f > tmpFile; mv tmpFile \$f;done
   run_ascat.r $bafTumor $logrTumor $bafNormal $logrNormal $idSampleTumor $baseDir
   """
 }

scripts/skeleton_batch.sh

@@ -21,11 +21,11 @@ ln -fs /castor/project/proj_nobackup/CAW/containers containers
 
 export NXF_TEMP=/scratch
 export NXF_LAUNCHBASE=/scratch
-export NXF_WORK=`pwd`/work
+export NXF_WORK=/scratch
 export NXF_HOME=/castor/project/proj_nobackup/nextflow
 export PATH=${NXF_HOME}/bin:${PATH}
-nextflow run ${CAW}/main.nf --step skipPreprocessing -profile singularityLocal --tools $2 --sample $1 -with-timeline ${PREFIX}.timeline.html -with-trace ${PREFIX}.trace.txt
+nextflow run ${CAW}/main.nf --step skipPreprocessing --tools $2 --sample $1 -with-timeline ${PREFIX}.timeline.html -with-trace ${PREFIX}.trace.txt
 
 # for annotation run
 # sbatch -J sample-Ann ./whatever.sh result.vcf[.gz] snpEff
-# nextflow run ${CAW}/main.nf --step annotate -profile singularityLocal --tools $2 --annotateVCF $1 --noReports --sample Preprocessing/Recalibrated/recalibrated.tsv -with-timeline ${PREFIX}.timeline.html -with-trace -with-trace ${PREFIX}.trace.txt
+# nextflow run ${CAW}/main.nf --step annotate --tools $2 --annotateVCF $1 --sample Preprocessing/Recalibrated/recalibrated.tsv -with-timeline ${PREFIX}.timeline.html -with-trace ${PREFIX}.trace.txt

scripts/speedseq.filter.awk

@@ -2,17 +2,22 @@ BEGIN{
 	MINQUAL=1;
 	SSC_THRES=-100;	# somatic score threshold ssc = LOD_T + LOD_N, log of odds (LOD) is the genotype quality ratio (http://www.nature.com/nmeth/journal/v12/n10/pdf/nmeth.3505.pdf)
 	ONLY_SOMATIC=1;	# prints out only somatic lines if not zero
-	NORMAL=10;	# index of normal sample genotype 
-	TUMOR=11; 	# index of tumor sample genotype
+	NORMAL=11;	# index of normal sample genotype 
+	TUMOR=10; 	# index of tumor sample genotype
 	GL_IDX=0;	# GL in the original: PL is the Normalized, Phred-scaled likelihoods for genotypes 
 }
 {
 	OFS="\t";
 
-    if ($0~"^#") { print ; next; }
+    if ($0~"^#" && $0!~"^#CHROM") { print ; next; }
+	# add extra header line
+	if ($0~"^#CHROM") { 
+			print "##INFO=<ID=DQUAL,Number=1,Type=Float,Description=\"SpeedSeq log of odds filter doi:10.1038/nmeth.3505 .\">"
+			print ; 
+	}
     if (! GL_IDX) {
         split($9,fmt,":")
-        for (i=1;i<=length(fmt);++i) { if (fmt[i]=="PL") GL_IDX=i }
+        for (i=1;i<=length(fmt);++i) { if (fmt[i]=="GL") GL_IDX=i }
     }
     split($NORMAL,N,":");		# split field 10 and put values into 
     split(N[GL_IDX],NGL,",");
@@ -29,11 +34,11 @@ BEGIN{
 
     if (DQUAL>=SSC_THRES && $NORMAL~"^0/0") {
         $7="PASS"
-        $8="SSC="DQUAL";"$8
+        $8="DQUAL="DQUAL";"$8
         print
     }
     else if (!ONLY_SOMATIC && $6>=MINQUAL && $10~"^0/0" && ! match($11,"^0/0")) {
-        $8="SSC="DQUAL";"$8
+        $8="DQUAL="DQUAL";"$8
         print
     }
 }