Merge pull request #40 from NAL-i5K/gff3_fix_phase

fix_phase - Set the initial phase as 0, if phase missing
NAL-i5K · Jun 18, 2018 · a1f3db5 · a1f3db5
2 parents 7f96313 + f536f3d
commit a1f3db5
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 4 deletions.
diff --git a/gff3tool/lib/gff3/gff3.py b/gff3tool/lib/gff3/gff3.py
@@ -291,7 +291,7 @@ def check_phase(self, initial_phase):
                     if cds_list[0]['phase'] != 0:
                         self.add_line_error(cds_list[0], {'message': '{0:s} {1:d}, should be {2:d}'.format(ERROR_INFO['Ema0006'], cds_list[0]['phase'], 0), 'error_type': 'PHASE', 'eCode': 'Ema0006'})
                     if type(cds_list[0]['phase']) != int:
-                        logger.warning('[Phase] Program failed. \n\t\t- Line {0:s}: {1:s}'.format(str(cds_list[0]['line_index']+1), cds_list[0]['line_raw']))
+                        logger.warning('[Phase] check_phase failed. \n\t\t- Line {0:s}: {1:s}'.format(str(cds_list[0]['line_index']+1), cds_list[0]['line_raw']))
                     continue
             strand = strand_set[0]
             if strand not in plus_minus:
@@ -309,13 +309,13 @@ def check_phase(self, initial_phase):
             else:
                 phase = sorted_cds_list[0]['phase']
                 if type(phase) != int:
-                    logger.warning('[Phase] Program failed. \n\t\t- Line {0:s}: {1:s}'.format(str(sorted_cds_list[0]['line_index']+1), sorted_cds_list[0]['line_raw']))
+                    logger.warning('[Phase] check_phase failed. \n\t\t- Line {0:s}: {1:s}'.format(str(sorted_cds_list[0]['line_index']+1), sorted_cds_list[0]['line_raw']))
             for line in sorted_cds_list:
                 if line['phase'] != phase:
                     try:
                         self.add_line_error(line, {'message': 'Wrong phase {0:d}, should be {1:d}'.format(line['phase'], phase), 'error_type': 'PHASE', 'eCode': 'Ema0006'})
                     except:
-                        logger.warning('[Phase] Program failed. \n\t\t- Line {0:s}: {1:s}'.format(str(line['line_index']+1), line['line_raw']))
+                        logger.warning('[Phase] check_phase failed. \n\t\t- Line {0:s}: {1:s}'.format(str(line['line_index']+1), line['line_raw']))
                 try:
                     phase = (3 - ((line['end'] - line['start'] + 1 - phase) % 3)) % 3
                 except:

diff --git a/gff3tool/lib/gff3_fix/fix.py b/gff3tool/lib/gff3_fix/fix.py
@@ -396,6 +396,7 @@ def fix_phase(gff3, error_list, line_num_dict, logger):
     Esf0027 : Phase is required for all CDS features
     """
     #valid_CDS_phase = set([0,1,2])
+    valid_phase = set((0, 1, 2))
     for error in error_list:
         for line_num in error:
             if gff3.lines[line_num-1]['line_status'] != 'removed':
@@ -420,7 +421,12 @@ def fix_phase(gff3, error_list, line_num_dict, logger):
                         if 'Ema0006' in line_num_dict[sorted_CDS_list[0]['line_index']+1]:
                             phase = map(int,re.findall(r'\d',line_num_dict[sorted_CDS_list[0]['line_index']+1]['Ema0006']))[1]
                         else:
-                            phase = sorted_CDS_list[0]['line_index']['phase']
+                            try:
+                                phase = sorted_CDS_list[0]['phase']
+                                if phase not in valid_phase:
+                                    phase = 0
+                            except ValueError:
+                                phase = 0
                         gff3.lines[sorted_CDS_list[0]['line_index']]['phase'] = phase
 
                     else: