Merge branch 'hmmscan-empty-results' into 'develop'

Hmmscan empty results

See merge request SOaAS/psot!4

scripts/convert_hmmer.py

@@ -9,6 +9,11 @@ parser.add_argument('--output', '-o', required=True, help='The converted results
 parser.add_argument('--dbxref', '-d', required=True, help='The dbxref prefix that will be prepended to the accession or id')
 args = parser.parse_args()
 
+query_file = args.result + "/queries.json"
+queries = []
+with open(query_file) as f:
+    queries = json.load(f)
+
 filename = args.result + "/domtblout.tsv"
 documents = {}
 tool = {'name': None, 'version': None, 'database': None}
@@ -30,7 +35,11 @@ with open(filename) as f:
             query_id = split[3]
             if not query_id in documents:
                 documents[query_id] = {"id": query_id, "computations": [{'tool': tool, 'results':[]}]}
-            results =  documents[query_id]['computations'][0]["results"]
+                try:
+                    queries.remove(query_id)
+                except ValueError:
+                    print('HMMER converter: Query ID "' + query_id + '" not found among initial queries!')
+            results = documents[query_id]['computations'][0]["results"]
 
             results.append({
                 'target': {
@@ -48,6 +57,10 @@ with open(filename) as f:
                     }
                 })
 
+for query_id in queries:
+    if not query_id in documents:
+        documents[query_id] = {"id": query_id, "computations": [{'tool': tool, 'results':[]}]}
+
 output_filename = args.output
 with open(output_filename, 'w') as o:
     json.dump(documents, o)

scripts/run_hmmer.py

 #!/usr/bin/env python3
 import env
 import argparse
+import json
 from psot import config
 from os import system, makedirs
 
@@ -17,8 +18,20 @@ makedirs(args.output, exist_ok=True)
 
 system(hmmscan_tool + 
         " -E " + args.evalue + 
-        " -o " +args.output+ "/hmmscan.out " + 
+        " -o " + args.output + "/hmmscan.out " + 
         " --tblout " + args.output + "/tblout.tsv "  + 
         " --domtblout " + args.output + "/domtblout.tsv " + 
         " --pfamtblout " + args.output + "/pfamtblout.tsv " + 
         args.database + " " +  args.fasta)
+
+# Provide a list of all query sequence names for conversion process
+queries = []
+
+with open(args.fasta) as f:
+    for line in f:
+        if line.startswith('>'):
+            queries.append(line.split()[0].strip().lstrip('>'))
+
+query_file = args.output + '/queries.json'
+with open(query_file, 'w') as o:
+    json.dump(queries, o)