#99 - Start storing GFF/GTF source

Author: davmlaw

generate_transcript_data/gff_parser.py

@@ -117,9 +117,14 @@ def _create_gene(feature, gene_accession):
         if biotype:
             biotype_set.add(biotype)
 
+        source_set = set()
+        if feature.source:
+            source_set.add(feature.source)
+
         return {
             "gene_symbol": gene_name,
             "biotype": biotype_set,
+            "source": source_set,
             "id": gene_accession,
             "description": description
         }
@@ -132,6 +137,7 @@ def _create_transcript(feature, transcript_accession, gene_data):
             "gene_version": gene_data.get("id"),
             "exons": [],
             "biotype": set(),
+            "source": set(),
             CONTIG: feature.iv.chrom,
             STRAND: feature.iv.strand,
         }
@@ -169,7 +175,6 @@ def _add_transcript_data(self, transcript_accession, transcript, feature):
         if note := feature.attr.get("Note"):
             transcript["note"] = note
 
-
     def _finish_process_features(self):
         for transcript_accession, transcript_data in self.transcript_data_by_accession.items():
             features_by_type = self.transcript_features_by_type.get(transcript_accession, {})
@@ -458,6 +463,10 @@ def handle_feature(self, feature):
                 gene_data["biotype"].add(biotype)
                 transcript["biotype"].add(biotype)
 
+            if feature.source:
+                gene_data["source"].add(feature.source)
+                transcript["source"].add(feature.source)
+
             self._handle_protein_version(transcript_accession, feature)
             self._add_tags_to_transcript_data(transcript, feature)
 
@@ -511,6 +520,9 @@ def handle_feature(self, feature):
                 if m := self.hgnc_pattern.match(description):
                     gene_data["hgnc"] = m.group(2)
 
+            if feature.source:
+                gene_data["source"].add(feature.source)
+
             self.gene_accession_by_feature_id[feature.attr["ID"]] = gene_accession
         else:
             transcript_accession = None
@@ -564,6 +576,8 @@ def handle_feature(self, feature):
                     elif feature.type not in EXCLUDE_BIOTYPES:
                         transcript["biotype"].add(feature.type)
 
+                    if feature.source:
+                        transcript["source"].add(feature.source)
 
 
     @staticmethod

generate_transcript_data/json_schema_version.py

@@ -6,5 +6,5 @@
 # 0.2.29 - Ensembl now has HGNC added from outside GTFs
 # 0.2.30 - Ensembl GRCh37 has canonical transcripts added from outside GTFs
 # 0.2.31 - Add 'metadata' - method/urls
-
-JSON_SCHEMA_VERSION = "0.2.31"
+# 0.2.32 - Add 'source' (GTF column #2) to build data
+JSON_SCHEMA_VERSION = "0.2.32"