Summary inconsistent bug

ariba/summary.py

@@ -16,7 +16,8 @@ def __init__(
       filenames=None,
       fofn=None,
       include_all_variant_columns=False,
-      min_id=90.0
+      min_id=90.0,
+      verbose=False,
     ):
         if filenames is None and fofn is None:
             raise Error('Error! Must supply filenames or fofn to Summary(). Cannot continue')
@@ -32,6 +33,7 @@ def __init__(
         self.include_all_variant_columns = include_all_variant_columns
         self.min_id = min_id
         self.outprefix = outprefix
+        self.verbose = verbose
 
 
     def _load_fofn(self, fofn):
@@ -48,11 +50,13 @@ def _check_files_exist(self):
 
 
     @classmethod
-    def _load_input_files(cls, filenames, min_id):
+    def _load_input_files(cls, filenames, min_id, verbose=False):
         samples = {}
         for filename in filenames:
             samples[filename] = summary_sample.SummarySample(filename, min_pc_id=min_id)
             samples[filename].run()
+            if verbose:
+                print('Loaded file', filename, flush=True)
         return samples
 
 
@@ -250,22 +254,45 @@ def _newick_from_dist_matrix(cls, distance_file, outfile):
 
 
     def run(self):
+        if self.verbose:
+            print('Loading input files...', flush=True)
         self._check_files_exist()
-        self.samples = self._load_input_files(self.filenames, self.min_id)
+        self.samples = self._load_input_files(self.filenames, self.min_id, verbose=self.verbose)
+        if self.verbose:
+            print('Generating output rows', flush=True)
         self.rows = self._gather_output_rows()
+
+        if self.verbose:
+            print('Filtering columns', flush=True)
         self.rows, remaining_clusters = Summary._filter_clusters(self.rows)
+
         if remaining_clusters == 0:
             print('No clusters found that are present in any sample. Will not write any output files', file=sys.stderr)
             sys.exit(1)
 
-        Summary._write_csv(self.filenames, self.rows, self.outprefix + '.csv', phandango=False)
+        csv_file = self.outprefix + '.csv'
+        if self.verbose:
+            print('Writing csv file', csv_file, flush=True)
+        Summary._write_csv(self.filenames, self.rows, csv_file, phandango=False)
 
         if len(self.samples) > 1:
-            lines = Summary._write_csv(self.filenames, self.rows, self.outprefix + '.phandango.csv', phandango=True)
+            if self.verbose:
+                print('Making Phandango csv file', csv_file, flush=True)
+            csv_file = self.outprefix + '.phandango.csv'
+            lines = Summary._write_csv(self.filenames, self.rows, csv_file, phandango=True)
             dist_matrix_file = self.outprefix + '.phandango.distance_matrix'
             tree_file = self.outprefix + '.phandango.tre'
+
+            if self.verbose:
+                print('Making Phandango distance matrix', dist_matrix_file, flush=True)
             Summary._write_distance_matrix(lines, dist_matrix_file)
+
+            if self.verbose:
+                print('Making Phandango tree file', tree_file, flush=True)
             Summary._newick_from_dist_matrix(dist_matrix_file, tree_file)
             os.unlink(dist_matrix_file)
         else:
             print('Made csv file. Not making Phandango files because only one input file given', file=sys.stderr)
+
+        if self.verbose:
+            print('Finished', flush=True)

ariba/summary_cluster.py

@@ -146,17 +146,19 @@ def _get_nonsynonymous_var(data_dict):
 
         if not has_nonsyn:
             return None
-        elif data_dict['known_var_change'] == data_dict['ref_ctg_change'] == '.':
-            raise Error('Unexpected data in ariba summary... \n' + str(data_dict) + '\n... known_var_change and ref_ctg_change both equal to ".", but has a non synonymous change. Something is inconsistent. Cannot continue')
+        elif data_dict['known_var_change'] == data_dict['ref_ctg_change'] == '.' == data_dict['ref_ctg_effect']:
+            raise Error('Unexpected data in ariba summary... \n' + str(data_dict) + '\n... known_var_change, ref_ctg_change, ref_ctg_effect all equal to ".", but has a non synonymous change. Something is inconsistent. Cannot continue')
         else:
             if '.' not in [data_dict['known_var_change'], data_dict['ref_ctg_change']] and \
               data_dict['known_var_change'] != data_dict['ref_ctg_change']:
                 raise Error('Unexpected data in ariba summary... \n' + str(data_dict) + '\n... known_var_change != ref_ctg_change. Cannot continue')
 
             if data_dict['known_var_change'] != '.':
                 return data_dict['known_var_change']
-            else:
+            elif data_dict['ref_ctg_change'] != '.':
                 return data_dict['ref_ctg_change']
+            else:
+                return data_dict['ref_ctg_effect']
 
 
     def column_summary_data(self):

ariba/tasks/summary.py

@@ -10,6 +10,7 @@ def run():
     parser.add_argument('--no_var_columns', action='store_true', help='Do not keep a column for every variant. Default is to include them')
     parser.add_argument('--min_id', type=float, help='Minimum percent identity cutoff to count as assembled [%(default)s]', default=90, metavar='FLOAT')
     parser.add_argument('--no_filter', action='store_true', help='Do not filter rows or columns of output that are all 0 (by default, they are removed from the output)')
+    parser.add_argument('--verbose', action='store_true', help='Be verbose')
     parser.add_argument('outprefix', help='Prefix of output files')
     parser.add_argument('infiles', nargs='*', help='Files to be summarised')
     options = parser.parse_args()
@@ -22,5 +23,6 @@ def run():
         filenames=options.infiles,
         include_all_variant_columns=(not options.no_var_columns),
         min_id=options.min_id,
+        verbose=options.verbose
     )
     s.run()

ariba/tests/summary_cluster_test.py

@@ -209,6 +209,13 @@ def test_get_nonsynonymous_var(self):
         d['known_var_change'] = '.'
         self.assertEqual('P43Q', summary_cluster.SummaryCluster._get_nonsynonymous_var(d))
 
+        d['ref_ctg_change'] = '.'
+        with self.assertRaises(summary_cluster.Error):
+            summary_cluster.SummaryCluster._get_nonsynonymous_var(d)
+
+        d['ref_ctg_effect'] = 'MULTIPLE'
+        self.assertEqual('MULTIPLE', summary_cluster.SummaryCluster._get_nonsynonymous_var(d))
+
 
     def test_column_summary_data(self):
         '''Test column_summary_data'''