Merge branch 'hotfix/1.0.2'

Author: keiranmraine

CHANGES.md

@@ -1,5 +1,13 @@
+### 1.0.2
+* Update BRASS to handle data that is _very_ quiet
+* Fix up the wrapper timings file so it is captured properly
+* Upgraded base image to [dockstore-cgpwxs:2.0.4](https://github.com/cancerit/dockstore-cgpwxs/releases/tag/2.0.4)
+  * To pick up changes in [cgpPindel:2.2.0](https://github.com/cancerit/cgpPindel/releases/tag/v2.2.0)
+* Remove ability to build `*.bas` files as not possible to have optional secondary files and we state expectation of [dockstore-cgpmap](https://github.com/cancerit/dockstore-cgpmap) as data source.
+* Moved some processing around to reduce cpu wastage.
+
 ### 1.0.1
-* Update dependencies (cgpPindel/ascatNgs) to reduce reliance on Capture::Tiny, apparent cause of some issues.
+* Update dependencies to reduce reliance on Capture::Tiny, apparent cause of some issues
 
 ### 1.0.0
 * Test data in `examples/analysis_config.local.json` moved to a non-expiring location.

Dockerfile

@@ -1,9 +1,9 @@
-FROM  quay.io/wtsicgp/dockstore-cgpwxs:2.0.3
+FROM  quay.io/wtsicgp/dockstore-cgpwxs:2.0.4
 
 MAINTAINER  keiranmraine@gmail.com
 
 LABEL uk.ac.sanger.cgp="Cancer Genome Project, Wellcome Trust Sanger Institute" \
-      version="1.0.1" \
+      version="1.0.2" \
       description="The CGP WGS pipeline for dockstore.org"
 
 USER  root

Dockstore.cwl

@@ -19,7 +19,7 @@ dct:creator:
 
 requirements:
   - class: DockerRequirement
-    dockerPull: "quay.io/wtsicgp/dockstore-cgpwgs:1.0.1"
+    dockerPull: "quay.io/wtsicgp/dockstore-cgpwgs:1.0.2"
 
 hints:
   - class: ResourceRequirement
@@ -144,4 +144,9 @@ outputs:
     outputBinding:
       glob: WGS_*_vs_*.timings.tar.gz
 
+  global_time:
+    type: File
+    outputBinding:
+      glob: WGS_*_vs_*.time
+
 baseCommand: ["/opt/wtsi-cgp/bin/ds-wrapper.pl"]

build/opt-build.sh

@@ -23,7 +23,7 @@ SRC_ASCAT="https://raw.githubusercontent.com/Crick-CancerGenomics/ascat/6d40e69a
 VER_GRASS="v2.1.0"
 
 ### BRASS
-VER_BRASS="v5.3.1"
+VER_BRASS="v5.3.2"
 SOURCE_BLAT="http://users.soe.ucsc.edu/~kent/src/blatSrc35.zip"
 SRC_FASTA36="https://github.com/wrpearson/fasta36/archive/v36.3.8d_13Apr16.tar.gz"
 

examples/analysis_config.local.json

@@ -38,6 +38,10 @@
     "path": "/tmp/timings_WGS.tar.gz",
     "class": "File"
   },
+  "global_time": {
+    "path": "/tmp/global_WGS.time",
+    "class": "File"
+  },
   "run_params": {
     "path": "/tmp/params_WGS.params",
     "class": "File"

scripts/analysisWGS.sh

@@ -114,6 +114,8 @@ ln -fs $BAM_MT $BAM_MT_TMP
 ln -fs $BAM_WT $BAM_WT_TMP
 ln -fs $BAM_MT.bai $BAM_MT_TMP.bai
 ln -fs $BAM_WT.bai $BAM_WT_TMP.bai
+ln -fs $BAM_MT.bas $BAM_MT_TMP.bas
+ln -fs $BAM_WT.bas $BAM_WT_TMP.bas
 
 ## Make fake copynumber so we can run early steps of caveman
 perl -alne 'print join(qq{\t},$F[0],0,$F[1],2);' < $REF_BASE/genome.fa.fai | tee $TMP/norm.cn.bed > $TMP/tum.cn.bed
@@ -158,22 +160,6 @@ fi
 
 echo "Setting up Parallel block 1"
 
-if [ ! -f "${BAM_MT}.bas" ]; then
-  echo -e "\t[Parallel block 1] BAS $NAME_MT added..."
-  do_parallel[bas_MT]="bam_stats -i $BAM_MT_TMP -o $BAM_MT_TMP.bas"
-else
-  ln -fs $BAM_MT.bas $BAM_MT_TMP.bas
-  echo '#PRE EXISTING $NAME_MT.bam.bas file found' > $OUTPUT_DIR/timings/${PROTOCOL}_${NAME_MT}_vs_${NAME_WT}.time.bas_MT
-fi
-
-if [ ! -f "${BAM_WT}.bas" ]; then
-  echo -e "\t[Parallel block 1] BAS $NAME_WT added..."
-  do_parallel[bas_WT]="bam_stats -i $BAM_WT_TMP -o $BAM_WT_TMP.bas"
-else
-  ln -fs $BAM_WT.bas $BAM_WT_TMP.bas
-  echo '#PRE EXISTING $NAME_WT.bam.bas file found' > $OUTPUT_DIR/timings/${PROTOCOL}_${NAME_MT}_vs_${NAME_WT}.time.bas_WT
-fi
-
 echo -e "\t[Parallel block 1] Genotype Check added..."
 do_parallel[geno]="nice -n 10 compareBamGenotypes.pl \
  -o $OUTPUT_DIR/${PROTOCOL}_${NAME_MT}_vs_${NAME_WT}/genotyped \
@@ -187,6 +173,25 @@ do_parallel[verify_WT]="nice -n 10 verifyBamHomChk.pl -d 25 \
   -b $BAM_WT_TMP \
   -j $OUTPUT_DIR/${PROTOCOL}_${NAME_WT}/contamination/result.json"
 
+echo -e "\t[Parallel block 1] cgpPindel input added..."
+do_parallel[cgpPindel_input]="pindel.pl \
+ -o $OUTPUT_DIR/${PROTOCOL}_${NAME_MT}_vs_${NAME_WT}/pindel \
+ -r $REF_BASE/genome.fa \
+ -t $BAM_MT_TMP \
+ -n $BAM_WT_TMP \
+ -s $REF_BASE/pindel/simpleRepeats.bed.gz \
+ -u $REF_BASE/pindel/pindel_np.gff3.gz \
+ -f $REF_BASE/pindel/${PROTOCOL}_Rules.lst \
+ -g $REF_BASE/vagrent/codingexon_regions.indel.bed.gz \
+ -st $PROTOCOL \
+ -as $ASSEMBLY \
+ -sp '$SPECIES' \
+ -e $PINDEL_EXCLUDE \
+ -b $REF_BASE/pindel/HiDepth.bed.gz \
+ -c $CPU \
+ -sf $REF_BASE/pindel/softRules.lst \
+ -p input"
+
 echo -e "\t[Parallel block 1] BB alleleCount added..."
 if [ ! -z ${SKIPBB+x} ]; then
   do_parallel[alleleCount]="echo 'BB allele count disabled by params'"

scripts/ds-wrapper.pl

@@ -67,6 +67,9 @@
   die "ERROR: Please define assembly, not found in [CR|B]AM headers.\n";
 }
 
+$opts{'mt_sm'} = sample_name_from_xam($opts{'t'});
+$opts{'wt_sm'} = sample_name_from_xam($opts{'n'});
+
 printf "Options loaded: \n%s\n",Dumper(\%opts);
 
 ## unpack the reference area:
@@ -110,8 +113,7 @@
 print $R_FH qq{options(bitmapType='cairo')\n};
 close $R_FH;
 
-make_path($ENV{HOME}.'/timings');
-my $cmd = "/usr/bin/time -o $ENV{HOME}/timings/analysisWGS.time -v /opt/wtsi-cgp/bin/analysisWGS.sh";
+my $cmd = sprintf '/usr/bin/time -o %s/WGS_%s_vs_%s.time -v /opt/wtsi-cgp/bin/analysisWGS.sh', $ENV{HOME}, $opts{'mt_sm'}, $opts{'wt_sm'};
 exec($cmd); # I will never return to the perl code
 
 sub add_species_flag_ini {
@@ -130,6 +132,25 @@ sub add_species_flag_ini {
   return $ini_out;
 }
 
+sub sample_name_from_xam {
+  my $xam = shift;
+  my $sm;
+  open my $SAM, '-|', "samtools view -H $xam" or die $!;
+  while(my $line = <$SAM>) {
+    next unless($line =~ m/^\@RG/);
+    chomp $line;
+    $line .= "\t"; # simplify matching
+    if($line =~ m/SM:([^\t]+)\t/) {
+      my $smtmp = $1;
+      if(defined $sm && $smtmp ne $sm) {
+        die "Conflicting SM values found in different RG headers";
+      }
+      $sm = $smtmp;
+    }
+  }
+  return $sm;
+}
+
 sub species_assembly_from_xam {
   my $xam = shift;
   my %assembly_set;