Use fallback validation to OLS for taxon ID

Author: bistline

ingest/validation/validate_metadata.py

@@ -476,8 +476,17 @@ def retrieve_label_and_synonyms(
     :param property_type: attribute type for term (string, array, boolean)
     """
     ontology_name = re.split("[_:]", ontology_id)[0].lower()
-    if ontology_is_local(ontology_name):
+    # use OLS fallback for taxon IDs since minified version is not authoritative
+    use_fallback = ontology_name == "ncbitaxon"
+    if ontology_is_local(ontology_name) and not use_fallback:
         return minified_reader.find_ontology_entry(ontology_name, ontology_id, property_name)
+    elif ontology_is_local(ontology_name) and use_fallback:
+        try:
+            return minified_reader.find_ontology_entry(ontology_name, ontology_id, property_name)
+        except ValueError:
+            return retriever.retrieve_ontology_term_label_and_synonyms(
+                ontology_id, property_name, convention, property_type
+            )
     else:
         return retriever.retrieve_ontology_term_label_and_synonyms(
             ontology_id, property_name, convention, property_type

tests/data/annotation/metadata/convention/valid_extended_taxon_v3.0.0.txt

@@ -0,0 +1,7 @@
+NAME	cell_type	cell_type__ontology_label	organism_age	disease	disease__ontology_label	species	species__ontology_label	geographical_region	geographical_region__ontology_label	library_preparation_protocol	library_preparation_protocol__ontology_label	organ	organ__ontology_label	sex	is_living	organism_age__unit	organism_age__unit_label	ethnicity__ontology_label	ethnicity	race	race__ontology_label	donor_id	biosample_id	biosample_type	preservation_method
+TYPE	group	group	numeric	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group	group
+BM01_16dpp_AAGCAGTGGTAT	CL_0000066	epithelial cell	31	MONDO_0000001	disease	NCBITaxon_197152	Cloeon dipterum	GAZ_00003181	Boston	EFO_0008919	Seq-Well	UBERON_0001913	milk	female	yes	UO_0000036	year	European ancestry	HANCESTRO_0005	NCIT_C43862	Norwegian	BM01	BM01_16dpp_r3	PrimaryBioSample_BodyFluid	Fresh
+BM01_16dpp_TAAGCAGTGGTA	CL_0000066	epithelial cell	31	MONDO_0000001	disease	NCBITaxon_197152	Cloeon dipterum	GAZ_00003181	Boston	EFO_0008919	Seq-Well	UBERON_0001913	milk	female	yes	UO_0000036	year	European ancestry	HANCESTRO_0005	NCIT_C43862	Norwegian	BM01	BM01_16dpp_r3	PrimaryBioSample_BodyFluid	Fresh
+BM01_16dpp_CTAAGCAGTGGT	CL_0000066	epithelial cell	31	MONDO_0000001	disease	NCBITaxon_197152	Cloeon dipterum	GAZ_00003181	Boston	EFO_0008919	Seq-Well	UBERON_0001913	milk	female	yes	UO_0000036	year	European ancestry	HANCESTRO_0005	NCIT_C126538	Scandinavian	BM01	BM01_16dpp_r3	PrimaryBioSample_BodyFluid	Fresh
+BM01_16dpp_CGGTAAACCATT	CL_0000066	epithelial cell	31	MONDO_0000001	disease	NCBITaxon_197152	Cloeon dipterum	GAZ_00003181	Boston	EFO_0008919	Seq-Well	UBERON_0001913	milk	female	yes	UO_0000036	year	European ancestry	HANCESTRO_0005	NCIT_C126538	Scandinavian	BM01	BM01_16dpp_r3	PrimaryBioSample_BodyFluid	Fresh
+BM01_16dpp_CCGAATTCACCG	CL_0000066	epithelial cell	31	PATO_0000001	quality	NCBITaxon_197152	Cloeon dipterum	GAZ_00003181	Boston	EFO_0008919	Seq-Well	UBERON_0001913	milk	female	yes	UO_0000036	year	European ancestry	HANCESTRO_0005	NCIT_C126538	Scandinavian	BM01	BM01_16dpp_r3	PrimaryBioSample_BodyFluid	Fresh

tests/test_validate_metadata.py

@@ -961,6 +961,22 @@ def test_minified_reader_validation(self):
         self.assertEqual('type 1 diabetes mellitus', entry['label'])
         self.assertIn('insulin dependent diabetes', entry['synonyms'])
 
+    def test_uses_fallback_validation(self):
+        """Non-standard species should use OLS fallback validation"""
+        args = (
+            "--convention ../schema/alexandria_convention/alexandria_convention_schema.json "
+            "../tests/data/annotation/metadata/convention/valid_extended_taxon_v3.0.0.txt"
+        )
+        metadata, convention = self.setup_metadata(args)
+        self.assertTrue(
+            metadata.validate_format(), "Valid metadata headers should not elicit error"
+        )
+        validate_input_metadata(metadata, convention)
+        self.assertFalse(
+            report_issues(metadata), "Valid ontology content should not elicit error"
+        )
+        self.teardown_metadata(metadata)
+
 
 if __name__ == "__main__":
     unittest.main()