Merge pull request #110 from broadinstitute/dp-fasta-handling

handle spaces in fasta headers

test/input/TestToolPicard/messy-headers.fasta

@@ -0,0 +1,43 @@
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-1
+CTGGAGCAAGACAAATGATGCCGGATCAGATCGAGTGATGGTATCACCCCTGGCTGTGAC
+ATGGTGGAATAGGAATGGACCAACAACAAGTACAATTCACTATCCAAAGGTATACAAAAC
+AAAGGGAGAGAAGGCTAATGTGCTAATTGGGCAAGGAGACGTGGTGTTGGTGATGAAACG
+GAAACGGGACTCTAGCATACTTACTGACAGCCAGACAGCGACCAAAAGAATTCGGATGGC
+CATCAATTAGTGTGGAATTGTTTAAAAACGACCTTGTTTCTACT
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-2
+CGAAAGCAGGCAAACCATTTGAATGGATGTCAATCCGACCTTACTCTTCTTGAAAGTTCC
+AGCGCAAAATGCCATAAGCACCACATTCCCGTATACTGGAGATCCTCCATACAGCCATGG
+ACGGATTAAGAAGGAGGAGTTTGCTGAGATCATGAAGATCTGTTCCACCATTGAAGAGCT
+CAGACGGCAGAAATAGTGAATTTAGCTTGTCCTTCATGAAAAAATGCCTTGTTTCTACT
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-3
+AAGCAGGTACTGATTCAAAATGGAAGACTTTGTGCGACAATGCTTCAATCCAATGATTGT
+CGAGCTTGCGGAAAAAGCAATGAAAGAACATGGGGAAGATCCGAAAATCGAGACAAACAA
+GATTAACGATCCCTGGGTTTTGCTTAATGCATCTTGGTTCAACTCCTTCCTCACACATGC
+ACTGAAATAGTTGTGGCAATGCTACTATTTGCTATCCATACTGTCCAAAAAAGTACCTTG
+TTTCTACT
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-4
+GCAGTGTAGCTGGATGGCTCCTCGGAAACCCAATGTGCGACGAATTCATCAGAGTGCCGG
+AATGGTCTTACATAGTGGAGCGGGCTAACCCAGCTAATGACCTCTGTTACCCAGGGAGCC
+TATCTTTATGGATGTGCTCCAATGGGTCGTTACAATGCAGAATTTGCATTTAGATTTATG
+AGCTCAGATTGTAGTTAAAAACACCCTTGTTTCT
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-5
+AAGCAGGGTAGATAATCACTCACTGAGTGACATCCACATCATGGCGTCTCAAGGCACCAA
+ATCTGTTGGAAGAATGGTTGGCGGAATCGGGAGATTCTACATACAGATGTGCACTGAGCT
+GGACGAAAAGGCAACGAACCCGATCGTGCCTTCCTTTGACATGAACAATGAAGGATCTTA
+TTTCTTCGGAGACAATGCAGAGGAGTATGACAATTAAAGAAAAATACCCTTGTTTCTACT
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-6
+CGGGCAATTCATCTCTTTGCCCTATTAGTGGGTGGGCAATATACAGTAAGGACAACGGTA
+TAAGAATTGGATCTAAGGGGGATGTGTTTGTTATAAGAGAACCATTCATCTCATGCTCCC
+GTGGTGTAAATAGTGACACTGTGGGTTGGTCTTGGCCAGACGGTGCTGAGTTGCCATTCA
+CCATTGACAAGTAGTTTGTTCAAAAAACTCCTTGTTTCTACT
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-7
+AAGCAGGTAGATATTGAAAGATGAGTCTTCTAACCGAGGTCGAAACGTACGTTCTCTCTA
+TCGTCCCGTCGGGCCCCCTCAAAGCCGAGATCGCGCAGAGACTTGAAGATGTCTTTGCAG
+GGAAGAACACCGATCTTGAGGCTCTCATGGAATGGCTAAAGACAAGACCAATCCTGTCAC
+CTCTGACCAAGGGGATTTTGGGATTTGTGTTCACGCTCACCGTGCCCAGTGAGCGAGGAC
+CT
+>Influenza A virus A/bovine_milk/USA/Broad_MC_023/2024(H5N1))-8
+AGCDGGGTGACAAAAACATAATGGATTCCAACACTGTGTTAAGCTTTCAGGTAGACTGCT
+TTCTTTGGCATGTCCGCAAACGATTTGCAGACCAAGAACTGGGTGATGCCCCATTCCTTG
+TCGAGACGGCCACTCGTGCTGGGAAGCAGATAGTGGAGAGGATTCTGGAGGAAGAATCCG
+TCGTTTCAGCTTATTTAATAATAAAAAACACCCTTGTTTCTACT

test/unit/test_tools_picard.py

@@ -36,6 +36,23 @@ def test_fasta_index(self):
                 actual_first3 = [re.sub(r'VN:1\.[4-6]','VN:1.4',x.strip()).split('\t')[:3] for x in inf.readlines()]
             self.assertEqual(actual_first3, expected_first3)
 
+    def test_messy_fasta_index(self):
+        orig_ref = os.path.join(util.file.get_test_input_path(self), 'messy-headers.fasta')
+        picard_index = tools.picard.CreateSequenceDictionaryTool()
+        with util.file.tempfname('.fasta') as inRef:
+            shutil.copyfile(orig_ref, inRef)
+            picard_index.execute(inRef, overwrite=True)
+        with open(inRef[:-6] + '.dict', 'rt') as inf:
+            seqnames = set()
+            for line in inf:
+                if line.startswith('@SQ'):
+                    seq_name = dict(x.split(':', maxsplit=1) for x in line.strip().split('\t')[1:])['SN']
+                    # old versions of code cut this off at "Influenza"
+                    self.assertGreater(len(seq_name), 50)
+                    seqnames.add(seq_name)
+            # require that all sequence names are unique
+            self.assertEqual(len(seqnames), 8)
+
     def test_sam_downsample(self):
         desired_count = 100
         tolerance = 0.02

util/file.py

@@ -651,7 +651,7 @@ def write_fasta_with_sanitized_ids(fasta_in, out_filepath):
         fasta_out = FastaIO.FastaWriter(handle, wrap=None)
         fasta_out.write_header()
         for record in SeqIO.parse(fasta_in, "fasta"):
-            record.id=sanitize_id_for_sam_rname(record.id)
+            record.id=sanitize_id_for_sam_rname(record.description)
             fasta_out.write_record(record)
     print("out_filepath",out_filepath)
     print("os.path.dirname(out_filepath)",os.path.dirname(out_filepath))