Fix bug: hmmer and ghostx don't include sequences with no results

scripts/convert_ghostx.py

@@ -4,6 +4,7 @@ import sys
 import json
 import argparse
 from os import path
+import glob
 import subprocess
 
 parser = argparse.ArgumentParser(description='Convert ghostx results to json documents')
@@ -55,6 +56,20 @@ with open(result_filename) as f:
 
         results.append(result)
 
+# add entries for files with no results
+queries = []
+path = "*.ids"
+for filename in glob.glob(path):
+    with open(filename) as f:
+        for line in f:
+            if line.startswith('>'):
+                queries.append(line.split()[0].strip().lstrip('>'))
+            else:
+                queries.append(line.split()[0].strip())
+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/convert_hmmer.py

@@ -12,12 +12,14 @@ args = parser.parse_args()
 
 # Provide a list of all query sequence names for conversion process
 queries = []
-path = "*_enum_headers.tsv"
+path = "*.ids"
 for filename in glob.glob(path):
     with open(filename) as f:
         for line in f:
             if line.startswith('>'):
                 queries.append(line.split()[0].strip().lstrip('>'))
+            else:
+                queries.append(line.split()[0].strip())
 
 filename = args.result + "/domtblout.tsv"
 documents = {}