norm: preserve canonical OBO prefix case in obo_handle (fixes ~503k dangling edges in monarch-kg)

tests/test_norm.py

@@ -86,3 +86,33 @@ def test_normalize_and_remove_extra_edges(self):
         with open(self.chebi_graph_edges, "r") as infile:
             afterlen = len(infile.readlines())
         self.assertLess(afterlen, beforelen, "Edges not removed?")
+
+
+class TestObohandle(TestCase):
+    """Tests for `obo_handle`'s prefix-casing behaviour."""
+
+    def test_uppercase_prefix_unchanged(self):
+        """Pure-uppercase OBO prefixes survive obo_handle as-is.
+
+        HP, MONDO, UBERON, CL, etc. were already uppercase, so the previous
+        `.upper()` was a no-op for them. Removing that call must not regress
+        the no-op.
+        """
+        from universalizer.norm import obo_handle
+        self.assertEqual(obo_handle("OBO:HP_0000118"), "HP:0000118")
+        self.assertEqual(obo_handle("OBO:MONDO_0000001"), "MONDO:0000001")
+        self.assertEqual(obo_handle("OBO:UBERON_0000061"), "UBERON:0000061")
+        self.assertEqual(obo_handle("OBO:CL_0000000"), "CL:0000000")
+
+    def test_mixed_case_prefix_preserved(self):
+        """Mixed-case OBO prefixes (FBbt, WBbt) keep their canonical case.
+
+        Regression test for the bug where `obo_handle` `.upper()`-ed the
+        prefix, turning `OBO:FBbt_00000001` into `FBBT:00000001` instead of
+        the canonical `FBbt:00000001`. This silently dropped every
+        Drosophila and C. elegans gene-expression edge in downstream
+        consumers (e.g. monarch-kg's alliance_gene_to_expression).
+        """
+        from universalizer.norm import obo_handle
+        self.assertEqual(obo_handle("OBO:FBbt_00000001"), "FBbt:00000001")
+        self.assertEqual(obo_handle("OBO:WBbt_0000001"), "WBbt:0000001")

universalizer/norm.py

@@ -502,10 +502,15 @@ def obo_handle(old_id: str) -> str:
         # Replace the first underscore with colon
         new_id = new_id.replace("_", ":", 1)
 
-        # Capitalize the prefix, but not the rest
-        # of the ID
-        split_id = new_id.split(":", 1)
-        new_id = f"{split_id[0].upper()}:{split_id[1]}"
+        # Preserve the canonical case of the prefix as it appears in the
+        # source OBO PURL — `FBbt`, `WBbt` etc. stay mixed-case; pure
+        # uppercase prefixes (HP, MONDO, UBERON, CL, ...) are unchanged.
+        # Previously this `.upper()`-ed the whole prefix, which broke any
+        # mixed-case OBO namespace by emitting e.g. `FBBT:` instead of the
+        # canonical `FBbt:`. That caused every Drosophila and C. elegans
+        # gene-expression edge in monarch-kg to dangle (~503k edges per
+        # release on `alliance_gene_to_expression_edges`), because Alliance
+        # was looking up nodes by the canonical mixed-case CURIE.
     else:
         new_id = old_id