Merge pull request #126 from mutalyzer/genbank-no-version

Parse genbank file without VERSION field

mutalyzer/parsers/genbank.py

@@ -441,8 +441,12 @@ class GBparser():
         #     the genbank file) are from the original NC reference. We try to
         #     set the .id field to the working value in the caller.
         record.source_id = biorecord.id
-        record.source_accession, record.source_version = biorecord.id.split('.')[:2]
-        record.source_gi = biorecord.annotations['gi']
+        try:
+            record.source_accession, record.source_version = biorecord.id.split('.')[:2]
+        except ValueError:
+            record.source_accession = biorecord.id
+            record.source_version = '1'
+        record.source_gi = biorecord.annotations.get('gi')
         record.organism = biorecord.annotations['organism']
 
         # Todo: This will change once we support protein references

tests/data/references.yml

@@ -144,6 +144,10 @@ MARK1:
     - null
   - - XM_005273136
     - null
+ADAC:
+  accession: UD_143772172095
+  checksum: 0b7f7991c1fb50bdfd04d3b0e405ecf3
+  filename: UD_143772172095.gb.bz2
 NG_008939.1:
   checksum: 114a03e16ad2f63531d796c2fb0d7039
   filename: NG_008939.1.gb.bz2

tests/test_parsers_genbank.py

@@ -48,3 +48,17 @@ def test_only_complete_genes_included(settings, references, parser):
     filename = os.path.join(settings.CACHE_DIR, '%s.gb.bz2' % accession)
     record = parser.create_record(filename)
     assert [g.name for g in record.geneList] == ['A1BG']
+
+@with_references('ADAC')
+def test_no_version(settings, references, parser):
+    """
+    Genbank file without 'version' field, so BioPython record.id is the
+    accession number without version. Our parser used to crash on that.
+
+    This genbank file was contributed by Gerard Schaafsma (original
+    source unknown).
+    """
+    accession = references[0].accession
+    genbank_filename = os.path.join(settings.CACHE_DIR,
+                                    '%s.gb.bz2' % accession)
+    parser.create_record(genbank_filename)