skip the empty genomes during download (fixes #32)

HadrienG · Nov 6, 2017 · 812ba9a · 812ba9a
1 parent 656112b
commit 812ba9a
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 1 deletion.
diff --git a/iss/download.py b/iss/download.py
@@ -47,10 +47,17 @@ def ncbi(kingdom, n_genomes):
                     retmode='txt')
                 try:
                     record = SeqIO.read(genome_record, 'fasta')
+                    n_count = record.seq.count('N') + record.seq.count('n')
+                    assert n_count / len(record) != 1.0
                 except http.client.IncompleteRead as e:
                     logger.warning(
                         'Failed to read downloaded genome. Skipping')
                     continue
+                except AssertionError as e:
+                    logger.warning(
+                        '%s only contains Ns. Skipping'
+                        % nucleotide_info['AccessionVersion'])
+                    continue
                 genomes.append(record)
                 n += 1
             else:

diff --git a/iss/generator.py b/iss/generator.py
@@ -67,6 +67,7 @@ def simulate_read(record, ErrorModel, i):
 
     EXPERIMENTAL. SHOULD BE MULTI-THREADABLE
     """
+    logger = logging.getLogger(__name__)
     sequence = record.seq
     header = record.id
 
@@ -78,7 +79,8 @@ def simulate_read(record, ErrorModel, i):
             0, len(record.seq) - (2 * read_length + insert_size))
     except ValueError as e:
         logger.error(
-            '%s too small for this ErrorModel:%s' % (record.id, e))
+            '%s shorter than template length for this ErrorModel:%s'
+            % (record.id, e))
         sys.exit(1)
 
     forward_end = forward_start + read_length