pyobo 0.10.6__tar.gz → 0.10.7__tar.gz

{pyobo-0.10.6/src/pyobo.egg-info → pyobo-0.10.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyobo
-Version: 0.10.6
+Version: 0.10.7
 Summary: Handling and writing OBO
 Home-page: https://github.com/pyobo/pyobo
 Download-URL: https://github.com/pyobo/pyobo/releases

{pyobo-0.10.6 → pyobo-0.10.7}/docs/source/conf.py

@@ -58,7 +58,7 @@ author = "Charles Tapley Hoyt"
 #
 
 # The full version, including alpha/beta/rc tags.
-release = "0.10.6"
+release = "0.10.7"
 
 # The short X.Y version.
 parsed_version = re.match(

{pyobo-0.10.6 → pyobo-0.10.7}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = pyobo
-version = 0.10.6
+version = 0.10.7
 description = Handling and writing OBO
 long_description = file: README.rst
 url = https://github.com/pyobo/pyobo

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/identifier_utils.py

@@ -119,7 +119,7 @@ def wrap_norm_prefix(f):
 
 def standardize_ec(ec: str) -> str:
     """Standardize an EC code identifier by removing all trailing dashes and dots."""
-    ec = ec.strip()
+    ec = ec.strip().replace(" ", "")
     for _ in range(4):
         ec = ec.rstrip("-").rstrip(".")
     return ec

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/obographs.py

@@ -19,7 +19,7 @@ from bioontologies.obograph import (
 from bioontologies.robot import ParseResults
 
 from pyobo.struct import Obo, Reference, Term
-from pyobo.struct.typedef import is_a
+from pyobo.struct.typedef import definition_source, is_a
 
 __all__ = [
     "graph_from_obo",
@@ -61,12 +61,12 @@ def _rewire(r: Reference) -> curies.Reference:
 
 
 def _get_class_node(term: Term) -> Node:
-    if not term.definition:
-        definition = None
-    else:
+    if term.definition or term.provenance:
         definition = Definition.from_parsed(
             value=term.definition, references=[_rewire(p) for p in term.provenance]
         )
+    else:
+        definition = None
 
     if term.xrefs:
         if not term.xref_types:
@@ -126,3 +126,11 @@ def _iter_edges(term: Term) -> Iterable[Edge]:
                 _rewire(typedef.reference),
                 _rewire(target),
             )
+
+    for provenance_reference in term.provenance:
+        yield Edge.from_parsed(
+            _rewire(term.reference),
+            _rewire(definition_source.reference),
+            _rewire(provenance_reference),
+        )
+    # TODO also look through xrefs and seealso to get provenance xrefs?

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/reader.py

@@ -444,20 +444,23 @@ def _extract_synonym(
             break
 
     stype: Optional[SynonymTypeDef] = None
-    if specificity is not None:  # go fishing for a synonym type definition
-        for _stype in synonym_typedefs.values():
-            # Since there aren't a lot of carefully defined synonym definitions, it
-            # can appear as a string or curie. Therefore, we might see temporary prefixes
-            # get added, so we should check against full curies as well as local unique
-            # identifiers
-            if rest.startswith(_stype.curie):
-                rest = rest[len(_stype.curie) :].strip()
-                stype = _stype
-                break
-            elif rest.startswith(_stype.identifier):
-                rest = rest[len(_stype.identifier) :].strip()
-                stype = _stype
-                break
+    for _stype in synonym_typedefs.values():
+        # Since there aren't a lot of carefully defined synonym definitions, it
+        # can appear as a string or curie. Therefore, we might see temporary prefixes
+        # get added, so we should check against full curies as well as local unique
+        # identifiers
+        if rest.startswith(_stype.curie):
+            rest = rest[len(_stype.curie) :].strip()
+            stype = _stype
+            break
+        elif rest.startswith(_stype.preferred_curie):
+            rest = rest[len(_stype.preferred_curie) :].strip()
+            stype = _stype
+            break
+        elif rest.startswith(_stype.identifier):
+            rest = rest[len(_stype.identifier) :].strip()
+            stype = _stype
+            break
 
     if not rest.startswith("[") or not rest.endswith("]"):
         logger.warning("[%s:%s] problem with synonym: %s", prefix, identifier, s)

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/sources/complexportal.py

@@ -82,6 +82,17 @@ def _parse_xrefs(s) -> List[Tuple[Reference, str]]:
             logger.warning("xref missing (: %s", xref)
             continue
         note = note.rstrip(")")
+        note.replace("rhea:rhea ", "rhea:")
+        note.replace("rhea:Rhea ", "rhea:")
+        note.replace("eccode::", "eccode:")
+        note.replace("eccode:EC:", "eccode:")
+        note.replace("eccode:RHEA:", "rhea:")
+        if note.lower().startswith("rhea "):
+            note = note[len("Rhea ") :]
+            if note.lower().startswith("rhea:rhea "):
+                note = note[len("rhea:rhea ") :]
+        if note.lower().startswith("EC:"):
+            note = note[len("EC:") :]
         try:
             reference = Reference.from_curie(xref_curie)
         except ValueError:

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/sources/geonames.py

@@ -38,7 +38,9 @@ def get_terms(*, force: bool = False) -> Collection[Term]:
     """Get terms."""
     code_to_country = get_code_to_country(force=force)
     code_to_admin1 = get_code_to_admin1(code_to_country, force=force)
-    code_to_admin2 = get_code_to_admin2(code_to_admin1, force=force)
+    code_to_admin2 = get_code_to_admin2(
+        code_to_country=code_to_country, code_to_admin1=code_to_admin1, force=force
+    )
     id_to_term = get_cities(
         code_to_country=code_to_country,
         code_to_admin1=code_to_admin1,
@@ -113,7 +115,7 @@ def get_code_to_admin1(
 
 
 def get_code_to_admin2(
-    code_to_admin1: Mapping[str, Term], *, force: bool = False
+    *, code_to_country: Mapping[str, Term], code_to_admin1: Mapping[str, Term], force: bool = False
 ) -> Mapping[str, Term]:
     """Get a mapping from admin2 code to term."""
     admin2_df = ensure_df(
@@ -134,8 +136,13 @@ def get_code_to_admin2(
         term.append_property("code", code)
         code_to_admin2[code] = term
         admin1_code = code.rsplit(".", 1)[0]
-        admin1_term = code_to_admin1[admin1_code]
-        term.append_relationship(part_of, admin1_term)
+        admin1_term = code_to_admin1.get(admin1_code)
+        if admin1_term:
+            term.append_relationship(part_of, admin1_term)
+        else:
+            country_code = admin1_code.split(".", 1)[0]
+            country_term = code_to_country[country_code]
+            term.append_relationship(part_of, country_term)
     return code_to_admin2
 
 
@@ -197,7 +204,10 @@ def get_cities(
                     term.append_synonym(synonym)
 
         if pd.isna(admin1):
-            tqdm.write(f"[geonames:{identifier}] missing admin 1 code for {name} ({country})")
+            # TODO try to annotate these directly onto countries
+            tqdm.write(
+                f"[geonames:{identifier}] {name}, a city in {country}, is missing admin 1 code"
+            )
             continue
 
         admin1_full = f"{country}.{admin1}"

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/sources/npass.py

@@ -72,7 +72,9 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
 
         # TODO check that the first is always the parent compound?
         if pd.notna(pubchem_compound_ids):
-            pubchem_compound_ids = pubchem_compound_ids.split(";")
+            pubchem_compound_ids = [
+                yy.strip() for xx in pubchem_compound_ids.split(";") for yy in xx.strip().split(",")
+            ]
             if len(pubchem_compound_ids) > 1:
                 logger.debug("multiple cids for %s: %s", identifier, pubchem_compound_ids)
             for pubchem_compound_id in pubchem_compound_ids:

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/sources/uniprot/uniprot.py

@@ -30,7 +30,7 @@ class UniProtGetter(Obo):
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""
-        yield from iter_terms(force=force, version=self._version_or_raise)
+        yield from iter_terms(version=self._version_or_raise)
 
 
 def get_obo(force: bool = False) -> Obo:
@@ -38,9 +38,9 @@ def get_obo(force: bool = False) -> Obo:
     return UniProtGetter(force=force)
 
 
-def iter_terms(version: Optional[str] = None, force: bool = False) -> Iterable[Term]:
+def iter_terms(version: Optional[str] = None) -> Iterable[Term]:
     """Iterate over UniProt Terms."""
-    with open_reader(ensure(version=version, force=force)) as reader:
+    with open_reader(ensure(version=version)) as reader:
         _ = next(reader)  # header
         for uniprot_id, name, taxonomy_id, _synonyms, ecs, pubmeds, pdbs in tqdm(
             reader, desc="Mapping UniProt", unit_scale=True
@@ -63,11 +63,11 @@ def iter_terms(version: Optional[str] = None, force: bool = False) -> Iterable[T
             yield term
 
 
-def ensure(version: Optional[str] = None, force: bool = False) -> Path:
+def ensure(version: Optional[str] = None) -> Path:
     """Ensure the reviewed uniprot names are available."""
     if version is None:
         version = bioversions.get_version("uniprot")
-    return RAW_MODULE.ensure(PREFIX, version, name="reviewed.tsv.gz", url=REVIEWED_URL, force=force)
+    return RAW_MODULE.ensure(PREFIX, version, name="reviewed.tsv.gz", url=REVIEWED_URL)
 
 
 if __name__ == "__main__":

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/struct/reference.py

@@ -31,11 +31,15 @@ class Reference(curies.Reference):
             raise ExpansionError(f"Unknown prefix: {v}")
         return norm_prefix
 
+    @property
+    def preferred_prefix(self) -> str:
+        """Get the preferred curie for this reference."""
+        return bioregistry.get_preferred_prefix(self.prefix) or self.prefix
+
     @property
     def preferred_curie(self) -> str:
         """Get the preferred curie for this reference."""
-        pp = bioregistry.get_preferred_prefix(self.prefix) or self.prefix
-        return f"{pp}:{self.identifier}"
+        return f"{self.preferred_prefix}:{self.identifier}"
 
     @root_validator(pre=True)
     def validate_identifier(cls, values):  # noqa
@@ -125,7 +129,7 @@ class Reference(curies.Reference):
         if identifier_lower.startswith(f"{self.prefix.lower()}:"):
             rv = identifier_lower
         else:
-            rv = f"{self.prefix}:{self._escaped_identifier}"
+            rv = f"{self.preferred_prefix}:{self._escaped_identifier}"
         if self.name:
             rv = f"{rv} ! {self.name}"
         return rv

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/struct/struct.py

@@ -384,9 +384,9 @@ class Term(Referenced):
     ) -> None:
         """Append a property."""
         if isinstance(prop, (Reference, Referenced)):
-            prop = prop.curie
+            prop = prop.preferred_curie
         if isinstance(value, (Reference, Referenced)):
-            value = value.curie
+            value = value.preferred_curie
         self.properties[prop].append(value)
 
     def _definition_fp(self) -> str:
@@ -423,11 +423,11 @@ class Term(Referenced):
             yield f"def: {self._definition_fp()}"
 
         for xref in sorted(self.xrefs, key=attrgetter("prefix", "identifier")):
-            yield f"xref: {xref.preferred_curie}"
+            yield f"xref: {xref}"  # __str__ bakes in the ! name
 
         parent_tag = "is_a" if self.type == "Term" else "instance_of"
         for parent in sorted(self.parents, key=attrgetter("prefix", "identifier")):
-            yield f"{parent_tag}: {parent.preferred_curie}"
+            yield f"{parent_tag}: {parent}"  # __str__ bakes in the ! name
 
         for typedef, references in sorted(self.relationships.items(), key=_sort_relations):
             if (not typedefs or typedef not in typedefs) and (

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/struct/typedef.py

@@ -50,7 +50,10 @@ __all__ = [
 
 @dataclass
 class TypeDef(Referenced):
-    """A type definition in OBO."""
+    """A type definition in OBO.
+
+    See the subsection of https://owlcollab.github.io/oboformat/doc/GO.format.obo-1_4.html#S.2.2.
+    """
 
     reference: Reference
     comment: Optional[str] = None
@@ -108,6 +111,8 @@ class TypeDef(Referenced):
             _chain = " ".join(link.preferred_curie for link in self.holds_over_chain)
             _names = " / ".join(link.name or "_" for link in self.holds_over_chain)
             yield f"holds_over_chain: {_chain} ! {_names}"
+        if self.inverse:
+            yield f"inverse_of: {self.inverse}"
 
     @classmethod
     def from_triple(cls, prefix: str, identifier: str, name: Optional[str] = None) -> "TypeDef":
@@ -276,6 +281,9 @@ alternative_term = TypeDef.from_triple(
 has_ontology_root_term = TypeDef.from_triple(
     prefix=IAO_PREFIX, identifier="0000700", name="has ontology root term"
 )
+definition_source = TypeDef.from_triple(
+    prefix=IAO_PREFIX, identifier="0000119", name="definition source"
+)
 has_dbxref = TypeDef.from_curie("oboInOwl:hasDbXref", name="has database cross-reference")
 
 editor_note = TypeDef.from_triple(prefix=IAO_PREFIX, identifier="0000116", name="editor note")

{pyobo-0.10.6 → pyobo-0.10.7}/src/pyobo/version.py

@@ -14,7 +14,7 @@ __all__ = [
     "get_git_hash",
 ]
 
-VERSION = "0.10.6"
+VERSION = "0.10.7"
 
 
 def get_git_hash() -> str:

{pyobo-0.10.6 → pyobo-0.10.7/src/pyobo.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyobo
-Version: 0.10.6
+Version: 0.10.7
 Summary: Handling and writing OBO
 Home-page: https://github.com/pyobo/pyobo
 Download-URL: https://github.com/pyobo/pyobo/releases

{pyobo-0.10.6 → pyobo-0.10.7}/tests/test_get.py

@@ -19,6 +19,7 @@ from pyobo.reader import (
     iterate_node_synonyms,
     iterate_node_xrefs,
 )
+from pyobo.struct.struct import acronym
 from tests.constants import TEST_CHEBI_OBO_PATH, chebi_patch
 
 
@@ -105,6 +106,7 @@ class TestParseObonet(unittest.TestCase):
         iupac_name = SynonymTypeDef.from_text("IUPAC NAME", lower=False)
         synoynym_typedefs = {
             "IUPAC_NAME": iupac_name,
+            acronym.curie: acronym,
         }
 
         for expected_synonym, text in [
@@ -137,6 +139,22 @@ class TestParseObonet(unittest.TestCase):
                 Synonym(name="LTEC I", specificity="EXACT"),
                 '"LTEC I" []',
             ),
+            (
+                Synonym(name="HAdV-A", specificity="BROAD", type=acronym),
+                '"HAdV-A" BROAD OMO:0003012 []',
+            ),
+            (
+                Synonym(name="HAdV-A", specificity="BROAD", type=acronym),
+                '"HAdV-A" BROAD omo:0003012 []',
+            ),
+            (
+                Synonym(name="HAdV-A", specificity="EXACT", type=acronym),
+                '"HAdV-A" OMO:0003012 []',
+            ),
+            (
+                Synonym(name="HAdV-A", specificity="EXACT", type=acronym),
+                '"HAdV-A" omo:0003012 []',
+            ),
         ]:
             with self.subTest(s=text):
                 actual_synonym = _extract_synonym(