#387 start of processing the ontologies JSON Lines file

ontologies_jsonl_to_kg_jsonl.py

@@ -0,0 +1,147 @@
+import argparse
+import kg2_util
+import json
+
+OWL_CLASS_TAG = "owl:Class"
+SUBCLASS_TAG = "rdfs:subClassOf"
+DESCRIPTION_TAG = "obo:IAO_0000115"
+XREF_TAG = "oboInOwl:hasDbXref"
+ID_TAG = "rdf:about"
+NAME_TAG = "rdfs:label"
+EXACT_MATCH_TAG = "skos:exactMatch"
+COMMENT_TAG = "rdfs:comment"
+
+TEXT_KEY = "ENTRY_TEXT"
+RESOURCE_KEY = "rdf:resource"
+
+OWL_SOURCE_KEY = "owl_source"
+
+KEYS_DICT = dict()
+
+COMMENT_PREFIX = "COMMENTS: "
+
+CLASSES_DICT = dict()
+
+def get_args():
+	arg_parser = argparse.ArgumentParser()
+	arg_parser.add_argument('--test', dest='test',
+							action="store_true", default=False)
+	arg_parser.add_argument('inputFile', type=str)
+	arg_parser.add_argument('outputFile', type=str)
+	return arg_parser.parse_args()
+
+def process_ontology_item(ontology_item):
+	source = ontology_item.get(OWL_SOURCE_KEY, str())
+	for owl_class in ontology_item.get(OWL_CLASS_TAG, list()):
+		# Typically genid classes which don't neatly map onto the KG2 schema
+		if ID_TAG not in owl_class:
+			continue
+		# TODO: MAP THIS HERE, since not all sources use same IRIs for the same nodes
+		node_id = owl_class.get(ID_TAG, str())
+
+		# Configure the name
+		name_list = [name.get(TEXT_KEY, None) for name in owl_class.get("rdfs:label", dict()) if TEXT_KEY in name]
+		if len(name_list) == 0:
+			continue
+
+		# Configure the description
+		description_list = list()
+		description_list += [description.get(TEXT_KEY, None) for description in owl_class.get("obo:IAO_0000115", list()) if (TEXT_KEY in description)]
+		description_list += [COMMENT_PREFIX + description.get(TEXT_KEY, str()) for description in owl_class.get("rdfs:comment", list()) if (TEXT_KEY in description)]
+		description_list += [description.get(TEXT_KEY, None) for description in owl_class.get("obo:UBPROP_0000001", list()) if (TEXT_KEY in description)]
+		description_list += [description.get(TEXT_KEY, None) for description in owl_class.get("obo:UBPROP_0000005", list()) if (TEXT_KEY in description)]
+		description_list += [description.get(TEXT_KEY, None) for description in owl_class.get("efo1:source_description", list()) if (TEXT_KEY in description)]
+
+		# Configure the biological sequence
+		has_biological_sequence = dict()
+		has_biological_sequence['formula'] = [biological_sequence.get(TEXT_KEY, None) for biological_sequence in owl_class.get("chebi:formula", list()) if TEXT_KEY in biological_sequence]
+		has_biological_sequence['smiles'] = [biological_sequence.get(TEXT_KEY, None) for biological_sequence in owl_class.get("chebi:smiles", list()) if TEXT_KEY in biological_sequence]
+		has_biological_sequence['inchi'] = [biological_sequence.get(TEXT_KEY, None) for biological_sequence in owl_class.get("chebi:inchi", list()) if TEXT_KEY in biological_sequence]
+		has_biological_sequence['inchikey'] = [biological_sequence.get(TEXT_KEY, None) for biological_sequence in owl_class.get("chebi:inchikey", list()) if TEXT_KEY in biological_sequence]
+
+		# Extract edge triples
+		edges_list = list()
+
+		for edge_type in ["obo:RO_0002175", "obo:RO_0002161", "obo:RO_0002604", "obo:RO_0002171", "obo:RO_0002174", "obo:RO_0002475", "obo:RO_0001900", "obo:RO_0004050"]:
+			for edge in owl_class.get(edge_type, list()):
+				if RESOURCE_KEY in edge:
+					edges_list.append((edge_type, edge.get(RESOURCE_KEY, None)))
+
+		for edge_type in ["oboInOwl:hasDbXref"]:
+			for edge in owl_class.get(edge_type, list()):
+				if TEXT_KEY in edge:
+					edges_list.append((edge_type, edge.get(TEXT_KEY, None)))
+
+		restriction_edges = list()
+		restriction_edges += [(edge, "rdfs:subClassOf") for edge in owl_class.get("rdfs:subClassOf", list())]
+		for equiv in owl_class.get("owl:equivalentClass", list()):
+			for mini_class in equiv.get("owl:Class", list()):
+				for edge in mini_class.get("owl:intersectionOf", list()):
+					restriction_edges.append((edge, "owl:equivalentClass"))
+
+		for (edge, general_edge_type) in restriction_edges:
+			for restriction in edge.get("owl:Restriction", list()):
+				edge_type = restriction.get("owl:onProperty", list())
+				edge_object = restriction.get("owl:someValuesFrom", list())
+				if len(edge_type) != 1:
+					assert len(edge_type) <= 1, edge 
+					continue
+				if len(edge_object) != 1:
+					assert len(edge_object) <= 1, edge
+					continue
+				edge_type = edge_type[0].get(RESOURCE_KEY, None)
+				edge_object = edge_object[0].get(RESOURCE_KEY, None)
+
+				if edge_type != None and edge_object != None:
+					edges_list.append((edge_type, edge_object))
+
+			if RESOURCE_KEY in edge:
+				edges_list.append((general_edge_type, edge.get(RESOURCE_KEY, None)))
+
+
+		# node_id = owl_class.get(ID_TAG, list())
+
+		# superclasses = [superclass.get(RESOURCE_KEY, str()) for superclass in owl_class.get(SUBCLASS_TAG, list())]
+
+		# # Also query for comments?
+		# # Descriptions appear to be additive in current KG2
+		# descriptions = owl_class.get(DESCRIPTION_TAG, list())
+		# assert len(descriptions) <= 1
+		# description = str()
+		# for element in descriptions:
+		# 	description += element[TEXT_KEY]
+
+		# xrefs = [xref[TEXT_KEY] for xref in owl_class.get(XREF_TAG, list())]
+		# for element in owl_class.get(XREF_TAG, list()):
+		# 	xrefs.append(element[TEXT_KEY])
+
+		# exact_matches = [exact_match[RESOURCE_KEY] for exact_match in owl_class.get(EXACT_MATCH_TAG, list())]
+
+		# names = owl_class.get(NAME_TAG, list())
+		# assert len(names) <= 1, ontology_item
+		# name = str()
+		# for element in names:
+		# 	name += element[TEXT_KEY]
+
+		# node = {"id": node_id, "superclasses": superclasses, "description": description, "xrefs": xrefs, "name": name, "exact_matches": exact_matches}
+
+		node = {"id": node_id, "description_list": description_list, "name": name_list, "source": source, "has_biological_sequence": has_biological_sequence, "edges": edges_list}
+		print(json.dumps(node, indent=4))
+
+
+if __name__ == '__main__':
+	args = get_args()
+	input_file_name = args.inputFile
+	output_file_name = args.outputFile
+
+	input_read_jsonlines_info = kg2_util.start_read_jsonlines(input_file_name)
+	input_data = input_read_jsonlines_info[0]
+
+	owl_class_count = 0
+	for ontology_item in input_data:
+		process_ontology_item(ontology_item)
+
+	# print("OWL Classes:", owl_class_count)
+	# for key in KEYS_DICT:
+	# 	KEYS_DICT[key] = KEYS_DICT[key] / owl_class_count
+	# print(json.dumps(KEYS_DICT, indent=4, sort_keys=True))