Enforce integer step counts for ancestor enrichment (#48)

Author: ubyndr

pandasaurus/query.py

@@ -192,12 +192,12 @@ def contextual_slim_enrichment(self, context: List[str]) -> pd.DataFrame:
 
         return self.enriched_df
 
-    def ancestor_enrichment(self, step_count: str) -> pd.DataFrame:
+    def ancestor_enrichment(self, step_count: int) -> pd.DataFrame:
         """
         Perform ancestor enrichment analysis with a specified number of hops.
 
         Args:
-            step_count (str): The number of hops to consider when enriching terms.
+            step_count (int): The number of hops to consider when enriching terms.
 
         Returns:
             pd.DataFrame: A DataFrame containing enriched terms and associated information.
@@ -212,6 +212,10 @@ def ancestor_enrichment(self, step_count: str) -> pd.DataFrame:
         includes more distant ancestors.
 
         """
+        if not isinstance(step_count, int):
+            raise TypeError("step_count must be an integer")
+        if step_count < 1:
+            raise ValueError("step_count must be a positive integer")
         source_list = [term.get_iri() for term in self._term_list]
         query_string = get_ancestor_enrichment_query(source_list, step_count)
         object_list = list(set(uri for res in run_sparql_query(query_string) for uri in res.values()))

pandasaurus/utils/sparql_queries.py

@@ -46,7 +46,12 @@ def get_label_query(term_iri_list: List[str]) -> str:
     )
 
 
-def get_ancestor_enrichment_query(seed_list: List[str], step_count) -> str:
+def get_ancestor_enrichment_query(seed_list: List[str], step_count: int) -> str:
+    """Build a SPARQL query that walks up to `step_count` ancestors for the seeds."""
+    if not isinstance(step_count, int):
+        raise TypeError("step_count must be an integer")
+    if step_count < 1:
+        raise ValueError("step_count must be a positive integer")
     query = "SELECT * WHERE { GRAPH <http://reasoner.renci.org/nonredundant> "
     query += f"{{ VALUES ?s {{{' '.join(seed_list)} }} "
     query += "?s rdfs:subClassOf ?o0. "

test/test_query.py

@@ -285,6 +285,32 @@ def test_ancestor_enrichment(mocker):
     assert expected_simple_df["o"].reset_index(drop=True).equals(df["o"].reset_index(drop=True))
 
 
+def test_ancestor_enrichment_requires_integer_step(mocker):
+    mocker.patch(
+        "pandasaurus.curie_validator.run_sparql_query",
+        side_effect=[
+            iter(get_enrichment_validate_curie_list_result()),
+            iter(get_enrichment_find_obsolete_terms_data()),
+        ],
+    )
+    q = Query(blood_and_immune_test_data)
+    with pytest.raises(TypeError):
+        q.ancestor_enrichment("2")
+
+
+def test_ancestor_enrichment_requires_positive_step(mocker):
+    mocker.patch(
+        "pandasaurus.curie_validator.run_sparql_query",
+        side_effect=[
+            iter(get_enrichment_validate_curie_list_result()),
+            iter(get_enrichment_find_obsolete_terms_data()),
+        ],
+    )
+    q = Query(blood_and_immune_test_data)
+    with pytest.raises(ValueError):
+        q.ancestor_enrichment(0)
+
+
 def test_synonym_lookup(mocker):
     q = Query(["CL:0000084", "CL:0000813", "CL:0000815", "CL:0000900"])
 

test/utils/test_sparql_queries.py

@@ -1,5 +1,7 @@
 from test.data.query_data import get_blood_and_immune_test_data
 
+import pytest
+
 from pandasaurus.utils.sparql_queries import (
     get_ancestor_enrichment_query,
     get_contextual_enrichment_query,
@@ -96,6 +98,14 @@ def test_get_ancestor_enrichment_query():
     assert "o7" not in get_ancestor_enrichment_query(term_iri_list, 7)
 
 
+def test_get_ancestor_enrichment_query_requires_positive_int():
+    term_iri_list = ["term1"]
+    with pytest.raises(TypeError):
+        get_ancestor_enrichment_query(term_iri_list, "3")
+    with pytest.raises(ValueError):
+        get_ancestor_enrichment_query(term_iri_list, 0)
+
+
 def test_get_synonym_query():
     term_iri_list = ["term1", "term2", "term3"]
     expected_query = (