pyobo 0.10.5__tar.gz → 0.10.7__tar.gz

{pyobo-0.10.5/src/pyobo.egg-info → pyobo-0.10.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyobo
-Version: 0.10.5
+Version: 0.10.7
 Summary: Handling and writing OBO
 Home-page: https://github.com/pyobo/pyobo
 Download-URL: https://github.com/pyobo/pyobo/releases
@@ -49,7 +49,7 @@ Requires-Dist: class_resolver
 Requires-Dist: psycopg2-binary
 Requires-Dist: drugbank_downloader
 Requires-Dist: chembl_downloader
-Requires-Dist: umls_downloader
+Requires-Dist: umls_downloader>=0.1.2
 Requires-Dist: typing_extensions
 Provides-Extra: tests
 Requires-Dist: coverage; extra == "tests"

{pyobo-0.10.5 → 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 = "0.10.5"
+release = "0.10.7"
 
 # The short X.Y version.
 parsed_version = re.match(

{pyobo-0.10.5 → pyobo-0.10.7}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = pyobo
-version = 0.10.5
+version = 0.10.7
 description = Handling and writing OBO
 long_description = file: README.rst
 url = https://github.com/pyobo/pyobo
@@ -56,7 +56,7 @@ install_requires =
 	psycopg2-binary
 	drugbank_downloader
 	chembl_downloader
-	umls_downloader
+	umls_downloader>=0.1.2
 	typing_extensions
 zip_safe = false
 include_package_data = True

{pyobo-0.10.5 → pyobo-0.10.7}/src/pyobo/__init__.py

@@ -25,6 +25,7 @@ from .api import (  # noqa: F401
     get_name,
     get_name_by_curie,
     get_name_id_mapping,
+    get_obsolete,
     get_primary_curie,
     get_primary_identifier,
     get_properties,

{pyobo-0.10.5 → pyobo-0.10.7}/src/pyobo/api/__init__.py

@@ -27,6 +27,7 @@ from .names import (  # noqa: F401
     get_name,
     get_name_by_curie,
     get_name_id_mapping,
+    get_obsolete,
     get_synonyms,
 )
 from .properties import (  # noqa: F401

{pyobo-0.10.5 → pyobo-0.10.7}/src/pyobo/api/names.py

@@ -24,6 +24,7 @@ __all__ = [
     "get_id_definition_mapping",
     "get_synonyms",
     "get_id_synonyms_mapping",
+    "get_obsolete",
 ]
 
 logger = logging.getLogger(__name__)
@@ -184,6 +185,26 @@ def get_id_definition_mapping(
     return _get_mapping()
 
 
+def get_obsolete(
+    prefix: str,
+    *,
+    force: bool = False,
+    strict: bool = False,
+    version: Optional[str] = None,
+) -> Set[str]:
+    """Get the set of obsolete local unique identifiers."""
+    if version is None:
+        version = get_version(prefix)
+    path = prefix_cache_join(prefix, name="obsolete.tsv", version=version)
+
+    @cached_collection(path=path, force=force)
+    def _get_obsolete() -> Set[str]:
+        ontology = get_ontology(prefix, force=force, strict=strict, version=version)
+        return ontology.get_obsolete()
+
+    return set(_get_obsolete())
+
+
 @wrap_norm_prefix
 def get_synonyms(prefix: str, identifier: str) -> Optional[List[str]]:
     """Get the synonyms for an entity."""

{pyobo-0.10.5 → pyobo-0.10.7}/src/pyobo/gilda_utils.py

@@ -2,7 +2,6 @@
 
 """PyOBO's Gilda utilities."""
 
-import itertools as itt
 import logging
 from typing import Iterable, List, Optional, Tuple, Type, Union
 
@@ -11,6 +10,7 @@ import gilda.api
 import gilda.term
 from gilda.grounder import Grounder
 from gilda.process import normalize
+from gilda.term import filter_out_duplicates
 from tqdm.auto import tqdm
 
 from pyobo import (
@@ -18,6 +18,7 @@ from pyobo import (
     get_id_species_mapping,
     get_id_synonyms_mapping,
     get_ids,
+    get_obsolete,
 )
 from pyobo.getters import NoBuild
 from pyobo.utils.io import multidict
@@ -31,32 +32,6 @@ __all__ = [
 logger = logging.getLogger(__name__)
 
 
-_STATUSES = {"curated": 1, "name": 2, "synonym": 3, "former_name": 4}
-
-
-def filter_out_duplicates(terms: List[gilda.term.Term]) -> List[gilda.term.Term]:
-    """Filter out duplicates."""
-    # TODO import from gilda.term import filter_out_duplicates when it gets moved,
-    #  see https://github.com/indralab/gilda/pull/103
-    logger.debug("filtering %d terms for uniqueness", len(terms))
-    new_terms: List[gilda.term.Term] = [
-        min(terms_group, key=_status_key)
-        for _, terms_group in itt.groupby(sorted(terms, key=_term_key), key=_term_key)
-    ]
-    # Re-sort the terms
-    new_terms = sorted(new_terms, key=lambda x: (x.text, x.db, x.id))
-    logger.debug("got %d unique terms.", len(new_terms))
-    return new_terms
-
-
-def _status_key(term: gilda.term.Term) -> int:
-    return _STATUSES[term.status]
-
-
-def _term_key(term: gilda.term.Term) -> Tuple[str, str, str]:
-    return term.db, term.id, term.text
-
-
 def iter_gilda_prediction_tuples(
     prefix: str,
     relation: str = "skos:exactMatch",
@@ -115,10 +90,12 @@ def normalize_identifier(prefix: str, identifier: str) -> str:
 
 def get_grounder(
     prefixes: Union[str, Iterable[str]],
+    *,
     unnamed: Optional[Iterable[str]] = None,
     grounder_cls: Optional[Type[Grounder]] = None,
     versions: Union[None, str, Iterable[Union[str, None]]] = None,
     strict: bool = True,
+    skip_obsolete: bool = False,
 ) -> Grounder:
     """Get a Gilda grounder for the given prefix(es)."""
     unnamed = set() if unnamed is None else set(unnamed)
@@ -140,7 +117,11 @@ def get_grounder(
         try:
             p_terms = list(
                 get_gilda_terms(
-                    prefix, identifiers_are_names=prefix in unnamed, version=version, strict=strict
+                    prefix,
+                    identifiers_are_names=prefix in unnamed,
+                    version=version,
+                    strict=strict,
+                    skip_obsolete=skip_obsolete,
                 )
             )
         except NoBuild:
@@ -155,26 +136,50 @@ def get_grounder(
         return grounder_cls(terms_dict)
 
 
+def _fast_term(
+    *,
+    text: str,
+    prefix: str,
+    identifier: str,
+    name: str,
+    status: str,
+    organism: Optional[str] = None,
+) -> gilda.term.Term:
+    return gilda.term.Term(
+        norm_text=normalize(text),
+        text=text,
+        db=prefix,
+        id=identifier,
+        entry_name=name,
+        status=status,
+        source=prefix,
+        organism=organism,
+    )
+
+
 def get_gilda_terms(
     prefix: str,
+    *,
     identifiers_are_names: bool = False,
     version: Optional[str] = None,
     strict: bool = True,
+    skip_obsolete: bool = False,
 ) -> Iterable[gilda.term.Term]:
     """Get gilda terms for the given namespace."""
     id_to_name = get_id_name_mapping(prefix, version=version, strict=strict)
     id_to_species = get_id_species_mapping(prefix, version=version, strict=strict)
+    obsoletes = get_obsolete(prefix, version=version, strict=strict) if skip_obsolete else set()
 
     it = tqdm(id_to_name.items(), desc=f"[{prefix}] mapping", unit_scale=True, unit="name")
     for identifier, name in it:
-        yield gilda.term.Term(
-            norm_text=normalize(name),
+        if identifier in obsoletes:
+            continue
+        yield _fast_term(
             text=name,
-            db=prefix,
-            id=identifier,
-            entry_name=name,
+            prefix=prefix,
+            identifier=identifier,
+            name=name,
             status="name",
-            source=prefix,
             organism=id_to_species.get(identifier),
         )
 
@@ -184,29 +189,31 @@ def get_gilda_terms(
             id_to_synonyms.items(), desc=f"[{prefix}] mapping", unit_scale=True, unit="synonym"
         )
         for identifier, synonyms in it:
+            if identifier in obsoletes:
+                continue
             name = id_to_name[identifier]
             for synonym in synonyms:
-                yield gilda.term.Term(
-                    norm_text=normalize(synonym),
+                if not synonym:
+                    continue
+                yield _fast_term(
                     text=synonym,
-                    db=prefix,
-                    id=identifier,
-                    entry_name=name,
+                    prefix=prefix,
+                    identifier=identifier,
+                    name=name,
                     status="synonym",
-                    source=prefix,
                     organism=id_to_species.get(identifier),
                 )
 
     if identifiers_are_names:
         it = tqdm(get_ids(prefix), desc=f"[{prefix}] mapping", unit_scale=True, unit="id")
         for identifier in it:
-            yield gilda.term.Term(
-                norm_text=normalize(identifier),
+            if identifier in obsoletes:
+                continue
+            yield _fast_term(
                 text=identifier,
-                db=prefix,
-                id=identifier,
-                entry_name=None,
-                status="identifier",
-                source=prefix,
+                prefix=prefix,
+                identifier=identifier,
+                name=identifier,
+                status="name",
                 organism=id_to_species.get(identifier),
             )

{pyobo-0.10.5 → 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-0.10.5 → 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-0.10.5 → 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-0.10.5 → pyobo-0.10.7}/src/pyobo/sources/__init__.py

@@ -20,6 +20,7 @@ from .drugcentral import DrugCentralGetter
 from .expasy import ExpasyGetter
 from .famplex import FamPlexGetter
 from .flybase import FlyBaseGetter
+from .geonames import GeonamesGetter
 from .gwascentral_phenotype import GWASCentralPhenotypeGetter
 from .gwascentral_study import GWASCentralStudyGetter
 from .hgnc import HGNCGetter
@@ -46,6 +47,7 @@ from .pubchem import PubChemCompoundGetter
 from .reactome import ReactomeGetter
 from .rgd import RGDGetter
 from .rhea import RheaGetter
+from .ror import RORGetter
 from .selventa import SCHEMGetter, SCOMPGetter, SDISGetter, SFAMGetter
 from .sgd import SGDGetter
 from .slm import SLMGetter
@@ -74,6 +76,7 @@ __all__ = [
     "FlyBaseGetter",
     "GWASCentralPhenotypeGetter",
     "GWASCentralStudyGetter",
+    "GeonamesGetter",
     "HGNCGetter",
     "HGNCGroupGetter",
     "ICD10Getter",
@@ -98,6 +101,7 @@ __all__ = [
     "PomBaseGetter",
     "PubChemCompoundGetter",
     "RGDGetter",
+    "RORGetter",
     "ReactomeGetter",
     "RheaGetter",
     "SCHEMGetter",

{pyobo-0.10.5 → pyobo-0.10.7}/src/pyobo/sources/cgnc.py

@@ -8,6 +8,7 @@ from typing import Iterable
 import pandas as pd
 
 from pyobo.struct import Obo, Reference, Term, from_species
+from pyobo.struct.typedef import exact_match
 from pyobo.utils.path import ensure_df
 
 __all__ = [
@@ -25,7 +26,7 @@ class CGNCGetter(Obo):
 
     ontology = PREFIX
     dynamic_version = True
-    typedefs = [from_species]
+    typedefs = [from_species, exact_match]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""

{pyobo-0.10.5 → pyobo-0.10.7}/src/pyobo/sources/chembl.py

@@ -12,7 +12,7 @@ from typing import Iterable
 import chembl_downloader
 
 from pyobo.struct import Obo, Reference, Term
-from pyobo.struct.typedef import has_inchi, has_smiles
+from pyobo.struct.typedef import exact_match, has_inchi, has_smiles
 
 __all__ = [
     "ChEMBLCompoundGetter",
@@ -45,6 +45,7 @@ class ChEMBLCompoundGetter(Obo):
 
     ontology = "chembl.compound"
     bioversions_key = "chembl"
+    typedefs = [exact_match]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""

{pyobo-0.10.5 → 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-0.10.5 → pyobo-0.10.7}/src/pyobo/sources/depmap.py

@@ -8,6 +8,7 @@ import pandas as pd
 import pystow
 
 from pyobo import Obo, Reference, Term
+from pyobo.struct.typedef import exact_match
 
 __all__ = [
     "get_obo",
@@ -23,6 +24,7 @@ class DepMapGetter(Obo):
 
     ontology = bioversions_key = PREFIX
     data_version = VERSION
+    typedefs = [exact_match]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""

{pyobo-0.10.5 → pyobo-0.10.7}/src/pyobo/sources/drugcentral.py

@@ -12,7 +12,7 @@ import psycopg2
 from tqdm.auto import tqdm
 
 from pyobo.struct import Obo, Reference, Synonym, Term
-from pyobo.struct.typedef import has_inchi, has_smiles
+from pyobo.struct.typedef import exact_match, has_inchi, has_smiles
 
 __all__ = [
     "DrugCentralGetter",
@@ -34,6 +34,7 @@ class DrugCentralGetter(Obo):
     """An ontology representation of the DrugCentral database."""
 
     ontology = bioversions_key = PREFIX
+    typedefs = [exact_match]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""