Merge pull request #77 from PacificBiosciences/bugfix/miniwdl-pbsv-split

Fix for concat_vcf for miniwdl. Link all inputs into a single directory so that vcfs and indices are together.
PacificBiosciences · Oct 18, 2023 · 49ddc79 · 49ddc79
2 parents ba7126d + b447c8e
commit 49ddc79
Show file tree

Hide file tree

Showing 4 changed files with 11 additions and 5 deletions.
diff --git a/wdl-ci.config.json b/wdl-ci.config.json
@@ -419,7 +419,7 @@
       "tasks": {
         "pbsv_call": {
           "key": "pbsv_call",
-          "digest": "77yon47d6t327ocrw6bed3dccyq5t3va",
+          "digest": "o5xv2etbm2j4s32d5xs626xj6sp2ykmj",
           "tests": [
             {
               "inputs": {
@@ -457,7 +457,7 @@
       "tasks": {
         "concat_vcf": {
           "key": "concat_vcf",
-          "digest": "ntfiawmetxbdacle2l7mpu5tkz2jmtz2",
+          "digest": "xkyvutmrg3gz6zgabdmwcjvcbwrbwwp7",
           "tests": [
             {
               "inputs": {

diff --git a/workflows/cohort_analysis/cohort_analysis.wdl b/workflows/cohort_analysis/cohort_analysis.wdl
@@ -32,7 +32,9 @@ workflow cohort_analysis {
 		File gvcf_index = gvcf_object.data_index
 	}
 
-	scatter (region_set in pbsv_splits) {
+	scatter (shard_index in range(length(pbsv_splits))) {
+        Array[String] region_set = pbsv_splits[shard_index]
+
 		call PbsvCall.pbsv_call {
 			input:
 				sample_id = cohort_id + ".joint",
@@ -41,6 +43,7 @@ workflow cohort_analysis {
 				reference = reference.fasta.data,
 				reference_index = reference.fasta.data_index,
 				reference_name = reference.name,
+				shard_index = shard_index,
 				regions = region_set,
 				mem_gb = pbsv_call_mem_gb,
 				runtime_attributes = default_runtime_attributes

diff --git a/workflows/sample_analysis/sample_analysis.wdl b/workflows/sample_analysis/sample_analysis.wdl
@@ -68,14 +68,17 @@ workflow sample_analysis {
 			runtime_attributes = default_runtime_attributes
 	}
 
-	scatter (region_set in pbsv_splits) {
+	scatter (shard_index in range(length(pbsv_splits))) {
+        Array[String] region_set = pbsv_splits[shard_index]
+
 		call PbsvCall.pbsv_call {
 			input:
 				sample_id = sample.sample_id,
 				svsigs = pbsv_discover.svsig,
 				reference = reference.fasta.data,
 				reference_index = reference.fasta.data_index,
 				reference_name = reference.name,
+				shard_index = shard_index,
 				regions = region_set,
 				runtime_attributes = default_runtime_attributes
 		}

diff --git a/workflows/wdl-common b/workflows/wdl-common
+8 −1		wdl/tasks/concat_vcf.wdl
+8 −5		wdl/tasks/pbsv_call.wdl
+0 −272		wdl/tasks/pharmcat.wdl
+28 −0		wdl/workflows/pharmcat/inputs.json
+271 −0		wdl/workflows/pharmcat/pharmcat.wdl