Merge pull request #141 from biowdl/biowdl-307

update samtools and htseq-count tasks for supporting more memory efficient counting (e.g. using name sorted bams)

CHANGELOG.md

@@ -34,3 +34,5 @@ version 1.0.0-dev
 + Add feature type, idattr and additional attributes to htseq-count.
 + Added allow-contain option to bowtie.
 + Added a changelog to keep track of changes.
++ Added sortByName task in samtools to support more memory efficient execution of HTSeqCount.
++ Removed the bam index from HTSeqCount's inputs.

htseq.wdl

@@ -3,7 +3,6 @@ version 1.0
 task HTSeqCount {
     input {
         Array[File]+ inputBams
-        Array[File]+ inputBamsIndex
         File gtfFile
         String outputTable = "output.tsv"
         String format = "bam"
@@ -40,4 +39,4 @@ task HTSeqCount {
         memory: memory
         docker: dockerImage
     }
-}
\ No newline at end of file
+}

samtools.wdl

@@ -88,6 +88,29 @@ task Merge {
     }
 }
 
+task SortByName {
+    input {
+        File bamFile
+        String outputBamPath = "namesorted.bam"
+
+        String dockerImage = "quay.io/biocontainers/samtools:1.8--h46bd0b3_5"
+    }
+
+    command {
+        set -e
+        mkdir -p $(dirname ~{outputBamPath})
+        samtools sort -n ~{bamFile} -o ~{outputBamPath}
+    }
+
+    output {
+        File outputBam = outputBamPath
+    }
+
+    runtime {
+        docker: dockerImage
+    }
+}
+
 task Markdup {
     input {
         File inputBam