feat: add support for mus musculus descendant organisms (#1248)

chanzuckerberg · Feb 7, 2025 · 0aaf700 · 0aaf700
1 parent 011bbad
commit 0aaf700
Show file tree

Hide file tree

Showing 4 changed files with 30 additions and 4 deletions.
diff --git a/cellxgene_schema_cli/cellxgene_schema/schema_definitions/schema_definition.yaml b/cellxgene_schema_cli/cellxgene_schema/schema_definitions/schema_definition.yaml
@@ -392,11 +392,11 @@ components:
           - # If organism is Mouse
             rule:
               column: organism_ontology_term_id
-              match_exact:
-                terms:
+              match_ancestors_inclusive:
+                ancestors:
                   - NCBITaxon:10090
             error_message_suffix: >-
-              When 'organism_ontology_term_id' is 'NCBITaxon:10090' (Mus musculus), 'development_stage_ontology_term_id' MUST be the most accurate descendant of 'MmusDv:0000001' or unknown.
+              When 'organism_ontology_term_id' is 'NCBITaxon:10090' (Mus musculus) or one of its descendants, 'development_stage_ontology_term_id' MUST be the most accurate descendant of 'MmusDv:0000001' or unknown.
             type: curie
             curie_constraints:
               ontologies:

diff --git a/cellxgene_schema_cli/cellxgene_schema/validate.py b/cellxgene_schema_cli/cellxgene_schema/validate.py
@@ -802,9 +802,16 @@ def _validate_column_dependencies(
             terms_to_match = set()
             column_to_match = dependency_def["rule"]["column"]
             if "match_ancestors_inclusive" in dependency_def["rule"]:
+                print("rule", dependency_def["rule"])
                 ancestors = dependency_def["rule"]["match_ancestors_inclusive"]["ancestors"]
                 for ancestor in ancestors:
+                    print("ancestor", ancestor)
+                    term_descendants = ONTOLOGY_PARSER.get_term_descendants(ancestor, include_self=True)
+                    print("term_descendants", term_descendants)
+                    print("len(terms to match) before", len(terms_to_match))
                     terms_to_match.update(ONTOLOGY_PARSER.get_term_descendants(ancestor, include_self=True))
+                    print("len(terms to match) after", len(terms_to_match))
+                print("terms to match", terms_to_match)
             if "match_exact" in dependency_def["rule"]:
                 terms_to_match.update(dependency_def["rule"]["match_exact"]["terms"])
             try:

diff --git a/cellxgene_schema_cli/requirements.txt b/cellxgene_schema_cli/requirements.txt
@@ -1,5 +1,5 @@
 anndata==0.11.2
-cellxgene-ontology-guide==1.4.1 # update before a schema migration
+cellxgene-ontology-guide==1.4.2 # update before a schema migration
 click<9
 Cython<4
 dask[array]==2024.12.0

diff --git a/cellxgene_schema_cli/tests/test_schema_compliance.py b/cellxgene_schema_cli/tests/test_schema_compliance.py
@@ -696,6 +696,25 @@ def test_cell_type_ontology_term_id(self, validator_with_adata, term):
         deprecated_error_message = f"ERROR: '{term}' in 'cell_type_ontology_term_id' is a deprecated term id of 'CL'."
         assert not_allowed_error_message in validator.errors or deprecated_error_message in validator.errors
 
+    @pytest.mark.parametrize(
+        "organism_ontology_term_id",
+        [
+            "NCBITaxon:10090",  # Mus musculus (ancestor term, always allowed)
+            "NCBITaxon:947985",  # Mus musculus albula
+            "NCBITaxon:35531",  # Mus musculus bactrianus
+        ],
+    )
+    def test_development_stage_ontology_term_id_mouse_descendant(self, validator_with_adata, organism_ontology_term_id):
+        """
+        If organism_ontolology_term_id is "NCBITaxon:10090" for Mus musculus OR its descendants,
+        this MUST be the most accurate MmusDv:0000001 descendant.
+        """
+        validator = validator_with_adata
+        obs = validator.adata.obs
+        obs.loc[obs.index[1], "organism_ontology_term_id"] = organism_ontology_term_id
+        validator.validate_adata()
+        assert not validator.errors
+
     @pytest.mark.parametrize(
         "development_stage_ontology_term_id,error",
         [