Typedef and xref cleanup

src/pyobo/sources/cgnc.py

@@ -72,9 +72,9 @@ def get_terms(force: bool = False) -> Iterable[Term]:
         )
         term.set_species(identifier="9031", name="Gallus gallus")
         if entrez_id and pd.notna(entrez_id):
-            term.append_xref(Reference(prefix="ncbigene", identifier=entrez_id))
+            term.append_exact_match(Reference(prefix="ncbigene", identifier=entrez_id))
         if pd.notna(ensembl_id):
-            term.append_xref(Reference(prefix="ensembl", identifier=ensembl_id))
+            term.append_exact_match(Reference(prefix="ensembl", identifier=ensembl_id))
         if synonym_1 and pd.notna(synonym_1):
             term.append_synonym(synonym_1)
         if synoynm_2 and pd.notna(synoynm_2):

src/pyobo/sources/chembl.py

@@ -12,6 +12,7 @@
 import chembl_downloader
 
 from pyobo.struct import Obo, Reference, Term
+from pyobo.struct.typedef import has_inchi, has_smiles
 
 __all__ = [
     "ChEMBLCompoundGetter",
@@ -66,11 +67,11 @@ def iter_terms(version: str) -> Iterable[Term]:
                 # TODO add xrefs?
                 term = Term.from_triple(prefix=PREFIX, identifier=chembl_id, name=name)
                 if smiles:
-                    term.append_property("smiles", smiles)
+                    term.append_property(has_smiles, smiles)
                 if inchi:
-                    term.append_property("inchi", inchi)
+                    term.append_property(has_inchi, inchi)
                 if inchi_key:
-                    term.append_xref(Reference(prefix="inchikey", identifier=inchi_key))
+                    term.append_exact_match(Reference(prefix="inchikey", identifier=inchi_key))
                 yield term
 
 

src/pyobo/sources/drugbank.py

@@ -17,7 +17,7 @@
 
 from ..getters import NoBuild
 from ..struct import Obo, Reference, Term
-from ..struct.typedef import has_salt
+from ..struct.typedef import has_inchi, has_salt, has_smiles
 from ..utils.cache import cached_pickle
 from ..utils.path import prefix_directory_join
 
@@ -121,10 +121,10 @@ def _make_term(drug_info: Mapping[str, Any]) -> Term:
         if identifier:
             term.append_xref(Reference(prefix=xref_prefix, identifier=identifier))
 
-    for prop in ["smiles", "inchi"]:
-        identifier = drug_info.get(xref_prefix)
+    for prop, debio_curie in [("smiles", has_smiles), ("inchi", has_inchi)]:
+        identifier = drug_info.get(prop)
         if identifier:
-            term.append_property(prop, identifier)
+            term.append_property(debio_curie, identifier)
 
     for salt in drug_info.get("salts", []):
         term.append_relationship(

src/pyobo/sources/drugcentral.py

@@ -12,6 +12,7 @@
 from tqdm.auto import tqdm
 
 from pyobo.struct import Obo, Reference, Synonym, Term
+from pyobo.struct.typedef import has_inchi, has_smiles
 
 __all__ = [
     "DrugCentralGetter",
@@ -87,13 +88,15 @@ def iter_terms() -> Iterable[Term]:
             xrefs=xrefs.get(drugcentral_id, []),
         )
         if inchi_key:
-            term.append_xref(Reference(prefix="inchikey", identifier=inchi_key))
-        term.append_property("smiles", smiles)
-        term.append_property("inchi", inchi)
+            term.append_exact_match(Reference(prefix="inchikey", identifier=inchi_key))
+        if smiles:
+            term.append_property(has_smiles, smiles)
+        if inchi:
+            term.append_property(has_inchi, inchi)
         if cas:
-            term.append_xref(Reference(prefix="cas", identifier=cas))
+            term.append_exact_match(Reference(prefix="cas", identifier=cas))
         yield term
 
 
 if __name__ == "__main__":
-    DrugCentralGetter.cli()
+    get_obo().write_default(write_obo=True)

src/pyobo/sources/hgncgenefamily.py

@@ -36,7 +36,7 @@ class HGNCGroupGetter(Obo):
     """An ontology representation of HGNC's gene group nomenclature."""
 
     ontology = PREFIX
-    dynamic_version = True
+    bioversions_key = "hgnc"
     synonym_typedefs = [symbol_type]
     typedefs = [from_species, enables]
 

src/pyobo/sources/mirbase_mature.py

@@ -39,15 +39,16 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
     for _, name, previous_name, mirbase_mature_id in tqdm(
         df.values, total=len(df.index), unit_scale=True
     ):
+        synonyms = []
+        if pd.notna(previous_name):
+            synonyms.append(Synonym(name=previous_name))
         yield Term(
             reference=Reference(
                 prefix=PREFIX, identifier=mirbase_mature_id, name=name if pd.notna(name) else None
             ),
-            synonyms=[
-                Synonym(name=previous_name),
-            ],
+            synonyms=synonyms,
         )
 
 
 if __name__ == "__main__":
-    MiRBaseMatureGetter.cli()
+    get_obo().write_default(write_obo=True, write_obograph=True, use_tqdm=True)

src/pyobo/sources/slm.py

@@ -7,7 +7,8 @@
 import pandas as pd
 from tqdm.auto import tqdm
 
-from pyobo import Obo, SynonymTypeDef, Term
+from pyobo import Obo, Reference, SynonymTypeDef, Term
+from pyobo.struct.typedef import has_inchi, has_smiles
 from pyobo.utils.path import ensure_df
 
 __all__ = [
@@ -90,22 +91,23 @@ def iter_terms(version: str, force: bool = False):
         else:
             raise ValueError(identifier)
         term = Term.from_triple(PREFIX, identifier, name)
-        term.append_property("level", level)
+        if pd.notna(level):
+            term.append_property("level", level)
         if pd.notna(abbreviation):
             term.append_synonym(abbreviation, type=abreviation_type)
         if pd.notna(synonyms):
             for synonym in synonyms.split("|"):
                 term.append_synonym(synonym.strip())
         if pd.notna(smiles):
-            term.append_property("smiles", smiles)
+            term.append_property(has_smiles, smiles)
         if pd.notna(inchi) and inchi != "InChI=none":
             if inchi.startswith("InChI="):
                 inchi = inchi[len("InChI=") :]
-            term.append_property("inchi", inchi)
+            term.append_property(has_inchi, inchi)
         if pd.notna(inchikey):
             if inchikey.startswith("InChIKey="):
                 inchikey = inchikey[len("InChIKey=") :]
-            term.append_property("inchikey", inchikey)
+            term.append_xref(Reference(prefix="inchikey", identifier=inchikey))
         if pd.notna(chebi_id):
             term.append_xref(("chebi", chebi_id))
         if pd.notna(lipidmaps_id):

src/pyobo/struct/reference.py

@@ -41,7 +41,7 @@ def validate_identifier(cls, values):  # noqa
         if norm_prefix is None:
             raise ExpansionError(f"Unknown prefix: {prefix}")
         values["prefix"] = norm_prefix
-        values["identifier"] = bioregistry.standardize_identifier(norm_prefix, identifier)
+        values["identifier"] = bioregistry.standardize_identifier(norm_prefix, identifier).strip()
         # if not bioregistry.is_valid_identifier(norm_prefix, values["identifier"]):
         #    raise ValueError(f"non-standard identifier: {norm_prefix}:{norm_identifier}")
         return values

src/pyobo/struct/struct.py

@@ -175,7 +175,7 @@ def _ensure_ref(reference: ReferenceHint) -> Reference:
         return Reference(prefix=reference[0], identifier=reference[1])
     if isinstance(reference, Reference):
         return reference
-    raise TypeError
+    raise TypeError(f"invalid type given for a reference ({type(reference)}): {reference}")
 
 
 @dataclass

src/pyobo/struct/typedef.py

@@ -40,6 +40,8 @@
     "enables",
     "participates_in",
     "has_participant",
+    "has_inchi",
+    "has_smiles",
 ]
 
 
@@ -292,6 +294,15 @@ def get_reference_tuple(relation: RelationHint) -> Tuple[str, str]:
     reference=Reference(prefix="ro", identifier="0018038", name="has functional parent"),
 )
 
+has_smiles = TypeDef(
+    reference=Reference(prefix="debio", identifier="0000022", name="has SMILES"),
+)
+
+has_inchi = TypeDef(
+    reference=Reference(prefix="debio", identifier="0000020", name="has InChI"),
+)
+
+
 default_typedefs: Dict[Tuple[str, str], TypeDef] = {
     v.pair: v for k, v in locals().items() if isinstance(v, TypeDef)
 }