Merge pull request #588 from broadinstitute/ct-parameterize-annot-transfer-aln-mem

align_and_annot_transfer_single: parameterize mem; set default mem higher 15GB->30GB; increase cpu count 4->8; strip other fasta extensions in genbank_single and impose regex on assembly_id pattern; map 'sus scrofa [domesticus]' to 'swine' in biosample_to_genbank

Author: tomkinsc

pipes/WDL/tasks/tasks_ncbi.wdl

@@ -181,6 +181,7 @@ task align_and_annot_transfer_single {
     Array[File]+ reference_feature_tables
 
     String       out_basename = basename(genome_fasta, '.fasta')
+    Int          machine_mem_gb = 30
     String       docker = "quay.io/broadinstitute/viral-phylo:2.4.1.0"
   }
 
@@ -222,8 +223,8 @@ task align_and_annot_transfer_single {
 
   runtime {
     docker: docker
-    memory: "15 GB"
-    cpu: 4
+    memory: machine_mem_gb + " GB"
+    cpu: 8
     dx_instance_type: "mem2_ssd1_v2_x4"
     preemptible: 1
     maxRetries: 2
@@ -954,8 +955,15 @@ task biosample_to_genbank {
               year = outrow['collection_date'].split('-')[0]
               country = outrow['geo_loc_name'].split(':')[0]
               host = outrow['host'].lower()
-              if host == 'homo sapiens':
-                host = 'human'
+
+              # host name mapping to the informal names used by NCBI in sequence titles for certain species
+              host_to_informal_name_map = {
+                'homo sapiens': 'human',
+                'sus scrofa domesticus': 'swine',
+                'sus scrofa': 'swine'
+              }
+              host = host_to_informal_name_map.get(host, host)
+
               if len(outrow['isolate'].split('/')) >= 2:
                 state_inst_labid = outrow['isolate'].split('/')[-2]
               else:

pipes/WDL/workflows/genbank_single.wdl

@@ -15,7 +15,8 @@ workflow genbank_single {
 
     input {
         File    assembly_fasta
-        String  assembly_id = basename(assembly_fasta, ".fasta")
+        String  assembly_id = basename(basename(basename(assembly_fasta, ".fasta"), ".fsa") , ".fa")
+        
         File?   aligned_bam
 
         String  ref_accessions_colon_delim
@@ -37,6 +38,10 @@ workflow genbank_single {
           description: "Genome to prepare for Genbank submission. All segments/chromosomes included in one file. Must contain exactly the same number of sequences as reference_accessions.",
           patterns: ["*.fasta"]
         }
+        assembly_id: {
+            description: "Unique identifier for this assembly. Defaults to the basename of assembly_fasta. table2asn requires this value to be <=50 characters; see: https://www.ncbi.nlm.nih.gov/genbank/table2asn/#fsa",
+            patterns: ["^[A-Za-z0-9\-_\.:\*#]{1,50}$"]
+        }
         ref_accessions_colon_delim: {
           description: "Reference genome Genbank accessions, each segment/chromosome in the exact same count and order as the segments/chromosomes described in assemblies_fasta. List of accessions should be colon delimited.",
           patterns: ["*.fasta"]