implement terminator export

Author: oschwengers

bakta/io/fasta.py

@@ -97,7 +97,7 @@ def write_ffn(features: Sequence[dict], ffn_path: Path):
     log.info('write feature nucleotide sequences: path=%s', ffn_path)
     with ffn_path.open('wt') as fh:
         for feat in features:
-            if(feat['type'] in [bc.FEATURE_T_RNA, bc.FEATURE_TM_RNA, bc.FEATURE_R_RNA, bc.FEATURE_NC_RNA, bc.FEATURE_NC_RNA_REGION, bc.FEATURE_CRISPR, bc.FEATURE_CDS, bc.FEATURE_SORF, bc.FEATURE_ORIC, bc.FEATURE_ORIV, bc.FEATURE_ORIT]):
+            if(feat['type'] in [bc.FEATURE_T_RNA, bc.FEATURE_TM_RNA, bc.FEATURE_R_RNA, bc.FEATURE_NC_RNA, bc.FEATURE_NC_RNA_REGION, bc.FEATURE_CRISPR, bc.FEATURE_CDS, bc.FEATURE_SORF, bc.FEATURE_TERMINATOR, bc.FEATURE_ORIC, bc.FEATURE_ORIV, bc.FEATURE_ORIT]):
                 identifier = feat['locus'] if 'locus' in feat else feat['id']
                 if(feat.get('product', '') != ''):
                     fh.write(f">{identifier} {feat['product']}\n{feat['nt']}\n")

bakta/io/gff.py

@@ -315,6 +315,19 @@ def write_features(data: dict, features_by_sequence: Dict[str, dict], gff3_path:
                     fh.write(f"{seq_id}\tBakta\t{so.SO_CDS.name}\t{start}\t{stop}\t.\t{feat['strand']}\t0\t{annotations}\n")
                     if(bc.FEATURE_SIGNAL_PEPTIDE in feat):
                         write_signal_peptide(fh, feat)
+                elif(feat['type'] == bc.FEATURE_TERMINATOR):
+                    annotations = {
+                        'ID': feat['id'],
+                        'Name': feat['product'],
+                        'product': feat['product'],
+                        'Dbxref': feat['db_xrefs']
+                    }
+                    if(cfg.compliant):
+                        annotations['inference'] = 'profile:TranstermHP:2.0'
+                        annotations['Dbxref'], annotations['Note'] = insdc.revise_dbxref_insdc(feat['db_xrefs'])  # remove INSDC invalid DbXrefs
+                        annotations[bc.INSDC_FEATURE_REGULATORY_CLASS] = insdc.select_regulatory_class(feat)
+                    annotations = encode_annotations(annotations)
+                    fh.write(f"{seq_id}\tTranstermHP\t{so.SO_REGULATORY_REGION.name}\t{start}\t{stop}\t.\t{feat['strand']}\t.\t{annotations}\n")
                 elif(feat['type'] == bc.FEATURE_GAP):
                     annotations = {
                         'ID': feat['id'],

bakta/io/insdc.py

@@ -215,6 +215,12 @@ def build_biopython_sequence_list(data: dict, features: Sequence[dict]):
                 insdc_feature_type = bc.INSDC_FEATURE_REGULATORY
                 qualifiers['note'].append(feature['product'])
                 qualifiers.pop('product', None)
+            elif(feature['type'] == bc.FEATURE_TERMINATOR):
+                qualifiers['inference'] = 'profile:TranstermHP:2.0'
+                qualifiers[bc.INSDC_FEATURE_REGULATORY_CLASS] = select_regulatory_class(feature)
+                insdc_feature_type = bc.INSDC_FEATURE_REGULATORY
+                qualifiers['note'].append(feature['product'])
+                qualifiers.pop('product', None)
             elif(feature['type'] == bc.FEATURE_CRISPR):
                 qualifiers[bc.INSDC_FEATURE_REPEAT_FAMILY] = 'CRISPR'
                 qualifiers[bc.INSDC_FEATURE_REPEAT_TYPE] = 'direct'
@@ -335,7 +341,9 @@ def select_regulatory_class(feature: dict) -> str:
             feature_class = so.SO(feature_class[0], feature_class[1])
             feature['class'] = feature_class
 
-        if(feature_class.id == so.SO_CIS_REG_ATTENUATOR.id):
+        if(feature_class.id == so.SO_CIS_REG_TERMINATOR.id):
+            return bc.INSDC_FEATURE_REGULATORY_CLASS_TERMINATOR
+        elif(feature_class.id == so.SO_CIS_REG_ATTENUATOR.id):
             return bc.INSDC_FEATURE_REGULATORY_CLASS_ATTENUATOR
         elif(feature_class.id == so.SO_CIS_REG_RIBOSWITCH.id):
             return bc.INSDC_FEATURE_REGULATORY_CLASS_RIBOSWITCH

bakta/io/tsv.py

@@ -133,8 +133,8 @@ def write_feature_inferences(sequences: Sequence[dict], features_by_sequence: Di
                         ])
                     )
                     fh.write('\n')
-                elif(feat['type'] in [bc.FEATURE_T_RNA, bc.FEATURE_R_RNA, bc.FEATURE_NC_RNA, bc.FEATURE_NC_RNA_REGION]):
-                    accession = '-' if feat['type'] == bc.FEATURE_T_RNA else [xref for xref in feat['db_xrefs'] if bc.DB_XREF_RFAM in xref][0]
+                elif(feat['type'] in [bc.FEATURE_T_RNA, bc.FEATURE_R_RNA, bc.FEATURE_NC_RNA, bc.FEATURE_NC_RNA_REGION, bc.FEATURE_TERMINATOR]):
+                    accession = '-' if feat['type'] in [bc.FEATURE_T_RNA, bc.FEATURE_TERMINATOR] else [xref for xref in feat['db_xrefs'] if bc.DB_XREF_RFAM in xref][0]
                     fh.write('\t'.join(
                         [
                             feat['sequence'] if 'sequence' in feat else feat['contig'],  # <1.10.0 compatibility