pyobo 0.12.10__py3-none-any.whl → 0.12.11__py3-none-any.whl

pyobo/sources/reactome.py

@@ -11,7 +11,7 @@ from tqdm.auto import tqdm
 from ..api import get_id_multirelations_mapping
 from ..constants import SPECIES_REMAPPING
 from ..resources.ncbitaxon import get_ncbitaxon_id
-from ..struct import Obo, Reference, Term, from_species, has_citation, has_participant
+from ..struct import Obo, Reference, Term, from_species, has_participant, is_mentioned_by
 from ..utils.io import multidict
 from ..utils.path import ensure_df
 
@@ -32,7 +32,7 @@ class ReactomeGetter(Obo):
     """An ontology representation of the Reactome pathway database."""
 
     ontology = bioversions_key = PREFIX
-    typedefs = [from_species, has_participant, has_citation]
+    typedefs = [from_species, has_participant, is_mentioned_by]
     root_terms = [ROOT]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
@@ -76,7 +76,7 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
             reference=Reference(prefix=PREFIX, identifier=reactome_id, name=name),
         )
         for pubmed_id in provenance_d.get(reactome_id, []):
-            term.append_provenance(Reference(prefix="pubmed", identifier=pubmed_id))
+            term.append_mentioned_by(Reference(prefix="pubmed", identifier=pubmed_id))
 
         if not taxonomy_id or pd.isna(taxonomy_id):
             raise ValueError(f"unmapped species: {species_name}")

pyobo/sources/rgd.py

@@ -9,22 +9,18 @@ from tqdm.auto import tqdm
 from pyobo.struct import (
     Obo,
     Reference,
-    SynonymTypeDef,
     Term,
-    default_reference,
     from_species,
-    has_citation,
     has_gene_product,
+    is_mentioned_by,
     transcribes_to,
 )
+from pyobo.struct.struct import previous_gene_symbol, previous_name
 from pyobo.utils.path import ensure_df
 
 logger = logging.getLogger(__name__)
 PREFIX = "rgd"
 
-old_symbol_type = SynonymTypeDef(reference=default_reference(PREFIX, "old_symbol"))
-old_name_type = SynonymTypeDef(reference=default_reference(PREFIX, "old_name"))
-
 # NOTE unigene id was discontinue in January 18th, 2021 dump
 
 GENES_URL = "https://download.rgd.mcw.edu/data_release/GENES_RAT.txt"
@@ -73,8 +69,8 @@ class RGDGetter(Obo):
     """An ontology representation of RGD's rat gene nomenclature."""
 
     bioversions_key = ontology = PREFIX
-    typedefs = [from_species, transcribes_to, has_gene_product, has_citation]
-    synonym_typedefs = [old_name_type, old_symbol_type]
+    typedefs = [from_species, transcribes_to, has_gene_product, is_mentioned_by]
+    synonym_typedefs = [previous_name, previous_gene_symbol]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""
@@ -119,11 +115,11 @@ def get_terms(force: bool = False, version: str | None = None) -> Iterable[Term]
         old_names = row["OLD_NAME"]
         if old_names and pd.notna(old_names):
             for old_name in old_names.split(";"):
-                term.append_synonym(old_name, type=old_name_type)
+                term.append_synonym(old_name, type=previous_name)
         old_symbols = row["OLD_SYMBOL"]
         if old_symbols and pd.notna(old_symbols):
             for old_symbol in old_symbols.split(";"):
-                term.append_synonym(old_symbol, type=old_symbol_type)
+                term.append_synonym(old_symbol, type=previous_gene_symbol)
         for prefix, key in namespace_to_column:
             xref_ids = str(row[key])
             if xref_ids and pd.notna(xref_ids):
@@ -154,7 +150,7 @@ def get_terms(force: bool = False, version: str | None = None) -> Iterable[Term]
         pubmed_ids = row["CURATED_REF_PUBMED_ID"]
         if pubmed_ids and pd.notna(pubmed_ids):
             for pubmed_id in str(pubmed_ids).split(";"):
-                term.append_provenance(Reference(prefix="pubmed", identifier=pubmed_id))
+                term.append_mentioned_by(Reference(prefix="pubmed", identifier=pubmed_id))
 
         term.set_species(identifier="10116", name="Rattus norvegicus")
         yield term

pyobo/sources/slm.py

@@ -7,7 +7,7 @@ from tqdm.auto import tqdm
 
 from pyobo import Obo, Reference, Term, TypeDef
 from pyobo.struct.struct import abbreviation as abbreviation_typedef
-from pyobo.struct.typedef import exact_match, has_citation, has_inchi, has_smiles
+from pyobo.struct.typedef import exact_match, has_inchi, has_smiles, is_mentioned_by
 from pyobo.utils.path import ensure_df
 
 __all__ = [
@@ -43,7 +43,7 @@ class SLMGetter(Obo):
     """An ontology representation of SwissLipid's lipid nomenclature."""
 
     ontology = bioversions_key = PREFIX
-    typedefs = [exact_match, LEVEL, has_inchi, has_smiles, has_citation]
+    typedefs = [exact_match, LEVEL, has_inchi, has_smiles, is_mentioned_by]
     synonym_typedefs = [abbreviation_typedef]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
@@ -117,7 +117,7 @@ def iter_terms(version: str, force: bool = False):
         for hmdb_id in _split(hmdb_ids):
             term.append_exact_match(("hmdb", hmdb_id))
         for pubmed_id in _split(pubmed_ids):
-            term.append_provenance(Reference(prefix="pubmed", identifier=pubmed_id))
+            term.append_mentioned_by(Reference(prefix="pubmed", identifier=pubmed_id))
         # TODO how to handle class, parents, and components?
         yield term
 

pyobo/sources/uniprot/uniprot.py

@@ -19,7 +19,7 @@ from pyobo.struct import (
     derives_from,
     enables,
     from_species,
-    has_citation,
+    is_mentioned_by,
     participates_in,
 )
 from pyobo.struct.typedef import gene_product_of, located_in, molecularly_interacts_with
@@ -68,7 +68,7 @@ class UniProtGetter(Obo):
         derives_from,
         located_in,
         IS_REVIEWED,
-        has_citation,
+        is_mentioned_by,
     ]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
@@ -156,7 +156,7 @@ def iter_terms(version: str | None = None) -> Iterable[Term]:
                         )
             for pubmed in pubmeds.split(";"):
                 if pubmed := pubmed.strip():
-                    term.append_provenance(Reference(prefix="pubmed", identifier=pubmed))
+                    term.append_mentioned_by(Reference(prefix="pubmed", identifier=pubmed))
             for pdb in pdbs.split(";"):
                 if pdb := pdb.strip():
                     term.append_xref(Reference(prefix="pdb", identifier=pdb))

pyobo/sources/wikipathways.py

@@ -80,7 +80,12 @@ def iter_terms(version: str, *, include_descriptions: bool = False) -> Iterable[
         taxonomy_name = SPECIES_REMAPPING.get(species_code, species_code)
 
         for identifier, _version, _revision, name, _species, genes in parse_wikipathways_gmt(path):
-            graph = read_zipfile_rdf(archive, inner_path=f"wp/{identifier}.ttl")
+            try:
+                graph = read_zipfile_rdf(archive, inner_path=f"wp/{identifier}.ttl")
+            except KeyError:
+                tqdm.write(f"[wikipathways:{identifier}] was not found inside zip file, skipping")
+                continue
+
             uri = f"https://identifiers.org/wikipathways/{identifier}"
 
             definition: str | None = None
@@ -122,7 +127,7 @@ def iter_terms(version: str, *, include_descriptions: bool = False) -> Iterable[
                     pw_references.add(ref)
                     term.append_parent(ref)
             if not parents:
-                tqdm.write(f"[{term.curie}] could not find parent")
+                tqdm.write(f"[{term.curie}] could not find annotation to parent in PW")
                 term.append_parent(ROOT)
 
             diseases = graph.query(

pyobo/struct/__init__.py

@@ -26,12 +26,12 @@ from .typedef import (
     from_species,
     gene_product_member_of,
     has_category,
-    has_citation,
     has_gene_product,
     has_member,
     has_part,
     has_participant,
     is_a,
+    is_mentioned_by,
     member_of,
     orthologous,
     part_of,
@@ -65,12 +65,12 @@ __all__ = [
     "from_species",
     "gene_product_member_of",
     "has_category",
-    "has_citation",
     "has_gene_product",
     "has_member",
     "has_part",
     "has_participant",
     "is_a",
+    "is_mentioned_by",
     "make_ad_hoc_ontology",
     "member_of",
     "orthologous",

pyobo/struct/functional/macros.py

@@ -72,7 +72,7 @@ class RelationshipMacro(Macro):
         annotations: f.Annotations | None = None,
     ) -> None:
         """Instantiate the object-to-object SubClassOf macro."""
-        super().__init__(f.SubClassOf(s, f.ObjectSomeValuesFrom(p, o)))
+        super().__init__(f.SubClassOf(s, f.ObjectSomeValuesFrom(p, o), annotations=annotations))
 
 
 class StringMacro(Macro):

pyobo/struct/functional/obo_to_functional.py

@@ -128,11 +128,15 @@ def get_term_axioms(term: Term) -> Iterable[f.Box]:
     if term.type == "Term":
         yield f.Declaration(s, type="Class")
         for parent in term.parents:
-            yield f.SubClassOf(s, parent)
-    else:
+            yield f.SubClassOf(s, parent, annotations=_get_annotations(term, pv.is_a, parent))
+    elif term.type == "Instance":
         yield f.Declaration(s, type="NamedIndividual")
         for parent in term.parents:
-            yield f.ClassAssertion(parent, s)
+            yield f.ClassAssertion(
+                parent, s, annotations=_get_annotations(term, pv.rdf_type, parent)
+            )
+    else:
+        raise ValueError(f"invalid term type: {term.type}")
     # 2
     if term.is_anonymous is not None:
         yield m.IsAnonymousMacro(s, term.is_anonymous)

pyobo/struct/obo/reader.py

@@ -668,7 +668,7 @@ def _handle_xref(
     # TODO this is not what spec calls for, maybe
     #  need a flag in macro config for this
     if xref.prefix in PROVENANCE_PREFIXES:
-        return term.append_provenance(xref, annotations=annotations)
+        return term.append_mentioned_by(xref, annotations=annotations)
 
     return term.append_xref(xref, annotations=annotations)
 

pyobo/struct/struct.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import datetime
+import inspect
 import itertools as itt
 import json
 import logging
@@ -106,6 +107,7 @@ SSSOM_DF_COLUMNS = [
     "contributor",
 ]
 FORMAT_VERSION = "1.4"
+_SOURCES = Path(__file__).parent.parent.joinpath("sources").resolve()
 
 
 @dataclass
@@ -246,12 +248,22 @@ DEFAULT_SYNONYM_TYPE = SynonymTypeDef(
     reference=Reference(prefix="oboInOwl", identifier="SynonymType", name="synonym type"),
 )
 abbreviation = SynonymTypeDef(
-    reference=Reference(prefix="OMO", identifier="0003000", name="abbreviation")
+    reference=Reference(prefix="omo", identifier="0003000", name="abbreviation")
 )
 acronym = SynonymTypeDef(reference=Reference(prefix="omo", identifier="0003012", name="acronym"))
 uk_spelling = SynonymTypeDef(
     reference=Reference(prefix="omo", identifier="0003005", name="UK spelling synonym")
 )
+previous_name = SynonymTypeDef(
+    reference=Reference(prefix="omo", identifier="0003008", name="previous name")
+)
+previous_gene_symbol = SynonymTypeDef(
+    reference=Reference(prefix="omo", identifier="0003015", name="previous gene symbol")
+)
+gene_symbol_synonym = SynonymTypeDef(
+    reference=Reference(prefix="omo", identifier="0003016", name="gene symbol synonym")
+)
+
 default_synonym_typedefs: dict[ReferenceTuple, SynonymTypeDef] = {
     abbreviation.pair: abbreviation,
     acronym.pair: acronym,
@@ -621,8 +633,15 @@ class Obo:
                 raise ValueError(f"{self.ontology} is missing data_version")
             elif "/" in self.data_version:
                 raise ValueError(f"{self.ontology} has a slash in version: {self.data_version}")
+
+        file_path = Path(inspect.getfile(self.__class__)).resolve()
+        script_url = f"https://github.com/biopragmatics/pyobo/blob/main/src/pyobo/sources/{file_path.relative_to(_SOURCES)}"
+
         if self.auto_generated_by is None:
-            self.auto_generated_by = f"PyOBO v{get_pyobo_version(with_git_hash=True)} on {datetime.datetime.now().isoformat()}"  # type:ignore
+            self.auto_generated_by = (
+                f"PyOBO v{get_pyobo_version(with_git_hash=True)} on "
+                f"{datetime.datetime.now().isoformat()} by {script_url}"
+            )  # type:ignore
 
     def _get_clean_idspaces(self) -> dict[str, str]:
         """Get normalized idspace dictionary."""
@@ -927,7 +946,9 @@ class Obo:
                 case OBOLiteral():
                     end = f'"{obo_escape_slim(value.value)}" {reference_escape(value.datatype, ontology_prefix=self.ontology)}'
                 case Reference():
-                    end = reference_escape(value, ontology_prefix=self.ontology)
+                    end = reference_escape(
+                        value, ontology_prefix=self.ontology, add_name_comment=True
+                    )
                 case _:
                     raise TypeError(f"Invalid property value: {value}")
             yield f"property_value: {reference_escape(predicate, ontology_prefix=self.ontology)} {end}"
@@ -946,21 +967,30 @@ class Obo:
                 license_literal = OBOLiteral.string(license_spdx_id)
             yield Annotation(v.has_license, license_literal)
 
-        if description := bioregistry.get_description(self.ontology):
-            yield Annotation(v.has_description, OBOLiteral.string(description.strip()))
-        if homepage := bioregistry.get_homepage(self.ontology):
-            yield Annotation(v.has_homepage, OBOLiteral.uri(homepage))
-        if repository := bioregistry.get_repository(self.ontology):
-            yield Annotation(v.has_repository, OBOLiteral.uri(repository))
-        if logo := bioregistry.get_logo(self.ontology):
-            yield Annotation(v.has_logo, OBOLiteral.uri(logo))
-        if mailing_list := bioregistry.get_mailing_list(self.ontology):
-            yield Annotation(v.has_mailing_list, OBOLiteral.string(mailing_list))
-        if (maintainer := bioregistry.get_contact(self.ontology)) and maintainer.orcid:
-            yield Annotation(
-                v.has_maintainer,
-                Reference(prefix="orcid", identifier=maintainer.orcid, name=maintainer.name),
-            )
+        if resource := bioregistry.get_resource(self.ontology):
+            if description := resource.get_description():
+                yield Annotation(v.has_description, OBOLiteral.string(description.strip()))
+            if homepage := resource.get_homepage():
+                yield Annotation(v.has_homepage, OBOLiteral.uri(homepage))
+            if repository := resource.get_repository():
+                yield Annotation(v.has_repository, OBOLiteral.uri(repository))
+            if logo := resource.get_logo():
+                yield Annotation(v.has_logo, OBOLiteral.uri(logo))
+            if mailing_list := resource.get_mailing_list():
+                yield Annotation(v.has_mailing_list, OBOLiteral.string(mailing_list))
+            if (maintainer := resource.get_contact()) and maintainer.orcid:
+                yield Annotation(
+                    v.has_maintainer,
+                    Reference(prefix="orcid", identifier=maintainer.orcid, name=maintainer.name),
+                )
+            for maintainer in resource.contact_extras or []:
+                if maintainer.orcid:
+                    yield Annotation(
+                        v.has_maintainer,
+                        Reference(
+                            prefix="orcid", identifier=maintainer.orcid, name=maintainer.name
+                        ),
+                    )
 
         # Root terms
         for root_term in self.root_terms or []:

pyobo/struct/struct_utils.py

@@ -5,6 +5,7 @@ from __future__ import annotations
 import datetime
 import itertools as itt
 import logging
+import warnings
 from abc import ABC, abstractmethod
 from collections import defaultdict
 from collections.abc import Iterable, Mapping, Sequence
@@ -840,7 +841,7 @@ class Stanza(Referenced, HasReferencesMixin):
         """Get definition provenance."""
         # return as a tuple to make sure nobody is appending on it
         return (
-            *self.get_property_objects(v.has_citation),
+            *self.get_property_objects(v.is_mentioned_by),
             # This gets all of the xrefs on _any_ axiom,
             # which includes the definition provenance
             *(
@@ -867,8 +868,18 @@ class Stanza(Referenced, HasReferencesMixin):
         *,
         annotations: Iterable[Annotation] | None = None,
     ) -> Self:
-        """Append a citation."""
-        return self.annotate_object(v.has_citation, reference, annotations=annotations)
+        """Append a creative work that mentions this term."""
+        warnings.warn("use append_mentioned_by instead", DeprecationWarning, stacklevel=2)
+        return self.append_mentioned_by(reference, annotations=annotations)
+
+    def append_mentioned_by(
+        self,
+        reference: Reference,
+        *,
+        annotations: Iterable[Annotation] | None = None,
+    ) -> Self:
+        """Append a creative work that mentions this term."""
+        return self.annotate_object(v.is_mentioned_by, reference, annotations=annotations)
 
 
 ReferenceHint: TypeAlias = (
@@ -1004,8 +1015,11 @@ def _format_obo_trailing_modifiers(
         match prop.value:
             case Reference():
                 right = reference_escape(prop.value, ontology_prefix=ontology_prefix)
-            case OBOLiteral(value, _datatype, _language):
-                right = value
+            case OBOLiteral(value, datatype, _language):
+                if datatype == v.xsd_string:
+                    right = f'"{obo_escape_slim(value)}"'
+                else:
+                    right = value
         modifiers.append((left, right))
     inner = ", ".join(f"{key}={value}" for key, value in modifiers)
     return " {" + inner + "}"

pyobo/struct/typedef.py

@@ -20,6 +20,7 @@ __all__ = [
     "derives_from_organism",
     "editor_note",
     "enables",
+    "ends",
     "exact_match",
     "example_of_usage",
     "from_species",
@@ -57,6 +58,7 @@ __all__ = [
     "role_of",
     "see_also",
     "species_specific",
+    "starts",
     "superclass_of",
     "transcribes_to",
     "translates_to",
@@ -116,6 +118,12 @@ molecularly_interacts_with = TypeDef(
 located_in = TypeDef(
     reference=Reference(prefix=RO_PREFIX, identifier="0001025", name="located in"),
 )
+starts = TypeDef(
+    reference=Reference(prefix=RO_PREFIX, identifier="0002223", name="starts"),
+)
+ends = TypeDef(
+    reference=Reference(prefix=RO_PREFIX, identifier="0002229", name="ends"),
+)
 contributes_to_condition = TypeDef(
     reference=Reference(prefix=RO_PREFIX, identifier="0003304", name="contributes to condition"),
 )
@@ -258,10 +266,15 @@ has_functional_parent = TypeDef(
     reference=Reference(prefix="ro", identifier="0018038", name="has functional parent"),
 )
 
-has_citation = TypeDef(
-    reference=v.has_citation,
+is_mentioned_by = TypeDef(
+    reference=v.is_mentioned_by,
+    is_metadata_tag=True,
+    inverse=v.mentions,
+)
+mentions = TypeDef(
+    reference=v.mentions,
     is_metadata_tag=True,
-    range=Reference(prefix="IAO", identifier="0000013", name="journal article"),
+    inverse=v.is_mentioned_by,
 )
 
 has_smiles = TypeDef(reference=v.has_smiles, is_metadata_tag=True).append_xref(v.debio_has_smiles)

pyobo/struct/vocabulary.py

@@ -5,7 +5,7 @@ from collections.abc import Sequence
 import curies
 from curies import vocabulary as _v
 
-from .reference import Reference, default_reference
+from .reference import Reference
 
 __all__ = [
     "equivalent_class",
@@ -84,8 +84,9 @@ has_inchi = Reference(prefix="chemrof", identifier="inchi_string")
 debio_has_smiles = Reference(prefix="debio", identifier="0000022", name="has SMILES")
 has_smiles = Reference(prefix="chemrof", identifier="smiles_string")
 
-# TODO update to use debio, or put in RO
-has_citation = default_reference(prefix="RO", identifier="hasCitation", name="has citation")
+is_mentioned_by = Reference(prefix="mito", identifier="isMentionedBy", name="is mentioned by")
+mentions = Reference(prefix="mito", identifier="mentions", name="mentions")
+
 has_description = _c(_v.has_description)
 has_license = _c(_v.has_license)
 has_title = _c(_v.has_title)

pyobo/utils/path.py

@@ -84,7 +84,7 @@ def ensure_df(
     dtype=str,
     verify: bool = True,
     backend: Literal["requests", "urllib"] | None = None,
-    **kwargs,
+    **kwargs: Any,
 ) -> pd.DataFrame:
     """Download a file and open as a dataframe."""
     _path = ensure_path(
@@ -107,12 +107,11 @@ def ensure_json(
     version: VersionHint = None,
     name: str | None = None,
     force: bool = False,
-    sep: str = "\t",
     dtype=str,
     verify: bool = True,
     backend: Literal["requests", "urllib"] | None = None,
-    **kwargs,
-) -> pd.DataFrame:
+    **kwargs: Any,
+) -> Any:
     """Download a file and open as JSON."""
     _path = ensure_path(
         prefix,
@@ -151,6 +150,8 @@ class CacheArtifact(enum.Enum):
     prefixes = "prefixes.json"
     metadata = "metadata.json"
 
+    embeddings = "embeddings.tsv.gz"
+
 
 def get_cache_path(
     ontology: str,

pyobo/version.py

@@ -12,7 +12,7 @@ __all__ = [
     "get_version",
 ]
 
-VERSION = "0.12.10"
+VERSION = "0.12.11"
 
 
 def get_git_hash() -> str:

{pyobo-0.12.10.dist-info → pyobo-0.12.11.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyobo
-Version: 0.12.10
+Version: 0.12.11
 Summary: A python package for handling and generating OBO
 Keywords: snekpack,cookiecutter,ontologies,biomedical ontologies,life sciences,natural sciences,bioinformatics,cheminformatics,Open Biomedical Ontologies,OBO
 Author: Charles Tapley Hoyt
@@ -55,10 +55,15 @@ Requires-Dist: nih-reporter-downloader>=0.0.1
 Requires-Dist: typing-extensions
 Requires-Dist: rdflib
 Requires-Dist: obographs>=0.0.8
+Requires-Dist: sssom-pydantic
 Requires-Dist: psycopg2-binary ; extra == 'drugcentral'
 Requires-Dist: ssslm[gilda] ; extra == 'gilda'
 Requires-Dist: ssslm[gilda-slim] ; extra == 'gilda-slim'
+Requires-Dist: grape ; extra == 'grape'
+Requires-Dist: embiggen ; extra == 'grape'
+Requires-Dist: ensmallen ; extra == 'grape'
 Requires-Dist: protmapper ; extra == 'pid'
+Requires-Dist: pykeen ; extra == 'pykeen'
 Requires-Dist: scispacy ; python_full_version < '3.13' and extra == 'scispacy'
 Requires-Dist: spacy ; python_full_version < '3.13' and extra == 'scispacy'
 Requires-Dist: scipy<1.11 ; python_full_version < '3.13' and extra == 'scispacy'
@@ -74,7 +79,9 @@ Project-URL: Repository, https://github.com/biopragmatics/pyobo.git
 Provides-Extra: drugcentral
 Provides-Extra: gilda
 Provides-Extra: gilda-slim
+Provides-Extra: grape
 Provides-Extra: pid
+Provides-Extra: pykeen
 Provides-Extra: scispacy
 Provides-Extra: sources
 Description-Content-Type: text/markdown