pyobo 0.12.1__tar.gz → 0.12.3__tar.gz

{pyobo-0.12.1 → pyobo-0.12.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyobo
-Version: 0.12.1
+Version: 0.12.3
 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
@@ -39,13 +39,14 @@ Requires-Dist: cachier
 Requires-Dist: pystow>=0.7.0
 Requires-Dist: bioversions>=0.8.0
 Requires-Dist: bioregistry>=0.12.7
-Requires-Dist: bioontologies>=0.7.1
+Requires-Dist: bioontologies>=0.7.2
 Requires-Dist: ssslm>=0.0.13
 Requires-Dist: zenodo-client>=0.3.6
 Requires-Dist: class-resolver>=0.6.0
 Requires-Dist: psycopg2-binary
 Requires-Dist: pydantic>=2.0
-Requires-Dist: curies>=0.10.13
+Requires-Dist: curies>=0.10.17
+Requires-Dist: curies-processing>=0.1.0
 Requires-Dist: python-dateutil
 Requires-Dist: networkx>=3.4
 Requires-Dist: drugbank-downloader
@@ -55,6 +56,7 @@ Requires-Dist: clinicaltrials-downloader>=0.0.2
 Requires-Dist: nih-reporter-downloader>=0.0.1
 Requires-Dist: typing-extensions
 Requires-Dist: rdflib
+Requires-Dist: obographs>=0.0.8
 Requires-Dist: ssslm[gilda] ; extra == 'gilda'
 Requires-Dist: ssslm[gilda-slim] ; extra == 'gilda-slim'
 Maintainer: Charles Tapley Hoyt
@@ -395,19 +397,6 @@ class MyTestCase(unittest.TestCase):
           pyobo.get_name("chebi", "1234")
 ```
 
-## Preprocessing CURIEs, URIs, and unqualified identifiers
-
-In order to normalize references and identify resources, PyOBO uses the
-[Bioregistry](https://github.com/bioregistry/bioregistry). It used to be a part
-of PyOBO, but has since been externalized for more general reuse.
-
-At
-[src/pyobo/identifier_utils/preprocessing.json](https://github.com/pyobo/pyobo/blob/master/src/pyobo/src/pyobo/identifier_utils/preprocessing.json)
-is the curated set of pre-processing rules. These are used in combination with
-the `curies` package to do pre-processing steps on CURIEs, URIs, and unqualified
-identifiers beyond what is possible with the Bioregistry. See
-https://curies.readthedocs.io/en/latest/preprocessing.html.
-
 ## Troubleshooting
 
 The OBO Foundry seems to be pretty unstable with respect to the URLs to OBO

{pyobo-0.12.1 → pyobo-0.12.3}/README.md

@@ -324,19 +324,6 @@ class MyTestCase(unittest.TestCase):
           pyobo.get_name("chebi", "1234")
 ```
 
-## Preprocessing CURIEs, URIs, and unqualified identifiers
-
-In order to normalize references and identify resources, PyOBO uses the
-[Bioregistry](https://github.com/bioregistry/bioregistry). It used to be a part
-of PyOBO, but has since been externalized for more general reuse.
-
-At
-[src/pyobo/identifier_utils/preprocessing.json](https://github.com/pyobo/pyobo/blob/master/src/pyobo/src/pyobo/identifier_utils/preprocessing.json)
-is the curated set of pre-processing rules. These are used in combination with
-the `curies` package to do pre-processing steps on CURIEs, URIs, and unqualified
-identifiers beyond what is possible with the Bioregistry. See
-https://curies.readthedocs.io/en/latest/preprocessing.html.
-
 ## Troubleshooting
 
 The OBO Foundry seems to be pretty unstable with respect to the URLs to OBO

{pyobo-0.12.1 → pyobo-0.12.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "pyobo"
-version = "0.12.1"
+version = "0.12.3"
 description = "A python package for handling and generating OBO"
 readme = "README.md"
 authors = [
@@ -72,13 +72,14 @@ dependencies = [
     "pystow>=0.7.0",
     "bioversions>=0.8.0",
     "bioregistry>=0.12.7",
-    "bioontologies>=0.7.1",
+    "bioontologies>=0.7.2",
     "ssslm>=0.0.13",
     "zenodo-client>=0.3.6",
     "class_resolver>=0.6.0",
     "psycopg2-binary",
     "pydantic>=2.0",
-    "curies>=0.10.13",
+    "curies>=0.10.17",
+    "curies-processing>=0.1.0",
     "python-dateutil",
     "networkx>=3.4",
     # Resource Downloaders
@@ -89,6 +90,7 @@ dependencies = [
     "nih-reporter-downloader>=0.0.1",
     "typing_extensions",
     "rdflib",
+    "obographs>=0.0.8",
 ]
 
 # see https://peps.python.org/pep-0735/ and https://docs.astral.sh/uv/concepts/dependencies/#dependency-groups
@@ -288,7 +290,7 @@ known-first-party = [
 docstring-code-format = true
 
 [tool.bumpversion]
-current_version = "0.12.1"
+current_version = "0.12.3"
 parse = "(?P<major>\\d+)\\.(?P<minor>\\d+)\\.(?P<patch>\\d+)(?:-(?P<release>[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*))?(?:\\+(?P<build>[0-9A-Za-z-]+(?:\\.[0-9A-Za-z-]+)*))?"
 serialize = [
     "{major}.{minor}.{patch}-{release}+{build}",

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/__init__.py

@@ -57,20 +57,29 @@ from .api import (
 )
 from .getters import get_ontology
 from .ner import get_grounder, ground
-from .obographs import parse_results_from_obo
 from .plugins import (
     has_nomenclature_plugin,
     iter_nomenclature_plugins,
     run_nomenclature_plugin,
 )
-from .reader import from_obo_path, from_obonet
-from .struct import Obo, Reference, Synonym, SynonymTypeDef, Term, TypeDef, default_reference
+from .struct import (
+    Obo,
+    Reference,
+    StanzaType,
+    Synonym,
+    SynonymTypeDef,
+    Term,
+    TypeDef,
+    default_reference,
+)
+from .struct.obo import from_obo_path, from_obonet
 from .utils.path import ensure_path
 from .version import get_version
 
 __all__ = [
     "Obo",
     "Reference",
+    "StanzaType",
     "Synonym",
     "SynonymTypeDef",
     "Term",
@@ -138,6 +147,5 @@ __all__ = [
     "has_nomenclature_plugin",
     "is_descendent",
     "iter_nomenclature_plugins",
-    "parse_results_from_obo",
     "run_nomenclature_plugin",
 ]

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/getters.py

@@ -20,7 +20,6 @@ from typing import Any, TypeVar
 import bioregistry
 import click
 import pystow.utils
-from bioontologies import robot
 from tabulate import tabulate
 from tqdm.auto import tqdm
 from typing_extensions import Unpack
@@ -34,8 +33,8 @@ from .constants import (
 )
 from .identifier_utils import ParseError, wrap_norm_prefix
 from .plugins import has_nomenclature_plugin, run_nomenclature_plugin
-from .reader import from_obo_path, from_obonet
 from .struct import Obo
+from .struct.obo import from_obo_path, from_obonet
 from .utils.io import safe_open_writer
 from .utils.path import ensure_path, prefix_directory_join
 from .version import get_git_hash, get_version
@@ -160,11 +159,20 @@ def get_ontology(
     elif ontology_format == "obo":
         pass  # all gucci
     elif ontology_format == "owl":
+        import bioontologies.robot
+
         _converted_obo_path = path.with_suffix(".obo")
         if prefix in REQUIRES_NO_ROBOT_CHECK:
             robot_check = False
-        robot.convert(path, _converted_obo_path, check=robot_check)
+        bioontologies.robot.convert(path, _converted_obo_path, check=robot_check)
         path = _converted_obo_path
+    elif ontology_format == "json":
+        from .struct.obograph import read_obograph
+
+        obo = read_obograph(prefix=prefix, path=path)
+        if cache:
+            obo.write_default(force=force_process)
+        return obo
     else:
         raise UnhandledFormatError(f"[{prefix}] unhandled ontology file format: {path.suffix}")
 

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/identifier_utils/__init__.py

@@ -1,5 +1,7 @@
 """Extract registry information."""
 
+from curies_processing import get_rules
+
 from .api import (
     DefaultCoercionError,
     EmptyStringError,
@@ -10,10 +12,10 @@ from .api import (
     UnregisteredPrefixError,
     _is_valid_identifier,
     _parse_str_or_curie_or_uri_helper,
+    get_converter,
     standardize_ec,
     wrap_norm_prefix,
 )
-from .preprocessing import get_rules
 from .relations import ground_relation
 
 __all__ = [
@@ -26,6 +28,7 @@ __all__ = [
     "UnregisteredPrefixError",
     "_is_valid_identifier",
     "_parse_str_or_curie_or_uri_helper",
+    "get_converter",
     "get_rules",
     "ground_relation",
     "standardize_ec",

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/identifier_utils/api.py

@@ -12,10 +12,10 @@ from bioregistry import NormalizedNamableReference as Reference
 from bioregistry.constants import FailureReturnType
 from curies import ReferenceTuple
 from curies.preprocessing import BlocklistError, PreprocessingConverter
+from curies_processing import get_rules
 from pydantic import ValidationError
 from typing_extensions import Doc
 
-from .preprocessing import get_rules
 from .relations import ground_relation
 
 __all__ = [
@@ -135,9 +135,10 @@ def _preclean_uri(s: str) -> str:
 
 
 @lru_cache(1)
-def _get_converter() -> PreprocessingConverter:
+def get_converter() -> PreprocessingConverter:
+    """Get a converter."""
     return PreprocessingConverter(
-        converter=bioregistry.manager.converter,
+        records=bioregistry.manager.converter.records,
         rules=get_rules(),
         preclean=_preclean_uri,
     )

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/sources/__init__.py

@@ -27,6 +27,7 @@ from .gtdb import GTDBGetter
 from .gwascentral import GWASCentralPhenotypeGetter, GWASCentralStudyGetter
 from .hgnc import HGNCGetter, HGNCGroupGetter
 from .icd import ICD10Getter, ICD11Getter
+from .intact import IntactGetter
 from .interpro import InterProGetter
 from .itis import ITISGetter
 from .kegg import KEGGGeneGetter, KEGGGenomeGetter, KEGGPathwayGetter
@@ -102,6 +103,7 @@ __all__ = [
     "ICD10Getter",
     "ICD11Getter",
     "ITISGetter",
+    "IntactGetter",
     "InterProGetter",
     "KEGGGeneGetter",
     "KEGGGenomeGetter",

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/sources/credit.py

@@ -7,7 +7,7 @@ from collections.abc import Iterable
 
 from more_itertools import chunked
 
-from pyobo.struct import Obo, Term
+from pyobo.struct import CHARLIE_TERM, HUMAN_TERM, Obo, Reference, Term, default_reference
 from pyobo.utils.path import ensure_path
 
 __all__ = [
@@ -16,6 +16,12 @@ __all__ = [
 
 url = "https://api.github.com/repos/CASRAI-CRedIT/Dictionary/contents/Picklists/Contributor%20Roles"
 PREFIX = "credit"
+ROOT = default_reference(prefix=PREFIX, identifier="contributor-role", name="contributor role")
+ROOT_TERM = (
+    Term(reference=ROOT)
+    .append_contributor(CHARLIE_TERM)
+    .append_xref(Reference(prefix="cro", identifier="0000000"))
+)
 
 
 class CreditGetter(Obo):
@@ -23,6 +29,7 @@ class CreditGetter(Obo):
 
     ontology = PREFIX
     static_version = "2022"
+    root_terms = [ROOT]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""
@@ -34,14 +41,16 @@ def get_terms(force: bool = False) -> list[Term]:
     path = ensure_path(PREFIX, url=url, name="picklist-api.json", force=force)
     with open(path) as f:
         data = json.load(f)
-    terms = []
+    terms = [
+        CHARLIE_TERM,
+        HUMAN_TERM,
+        ROOT_TERM,
+    ]
     for x in data:
-        name = x["name"].removesuffix(".md").lower()
-
         pp = ensure_path(PREFIX, "picklist", url=x["download_url"], backend="requests")
         with open(pp) as f:
             header, *rest = f.read().splitlines()
-            name = header = header.removeprefix("# Contributor Roles/")
+            name = header.removeprefix("# Contributor Roles/")
             dd = {k.removeprefix("## "): v for k, v in chunked(rest, 2)}
             identifier = (
                 dd["Canonical URL"]
@@ -50,7 +59,9 @@ def get_terms(force: bool = False) -> list[Term]:
             )
             desc = dd["Short definition"]
             terms.append(
-                Term.from_triple(prefix=PREFIX, identifier=identifier, name=name, definition=desc)
+                Term.from_triple(
+                    prefix=PREFIX, identifier=identifier, name=name, definition=desc
+                ).append_parent(ROOT)
             )
 
     return terms

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/sources/drugbank/drugbank.py

@@ -149,7 +149,7 @@ def get_xml_root(version: str | None = None) -> ElementTree.Element:
         raise NoBuildError from e
 
     element = parse_drugbank(version=version, username=username, password=password)
-    return element.getroot()
+    return element.getroot()  # type:ignore
 
 
 ns = "{http://www.drugbank.ca}"

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/sources/gwascentral/gwascentral_study.py

@@ -79,7 +79,7 @@ def iterate_terms(version: str, force: bool = False) -> Iterable[Term]:
                 except ElementTree.ParseError:
                     logger.warning("malformed XML in %s", tar_info.path)
                     continue
-            yield _get_term_from_tree(tree)
+            yield _get_term_from_tree(tree)  # type:ignore
 
 
 if __name__ == "__main__":

pyobo-0.12.3/src/pyobo/sources/intact.py

@@ -0,0 +1,79 @@
+"""Converter for IntAct complexes."""
+
+from __future__ import annotations
+
+from collections import defaultdict
+from collections.abc import Iterable
+
+import pandas as pd
+from pydantic import ValidationError
+from tqdm import tqdm
+
+from pyobo.struct import Obo, Reference, Term
+from pyobo.utils.path import ensure_path
+
+__all__ = [
+    "IntactGetter",
+]
+
+PREFIX = "intact"
+COMPLEXPORTAL_MAPPINGS_UNVERSIONED = (
+    "https://ftp.ebi.ac.uk/pub/databases/intact/complex/current/various/cpx_ebi_ac_translation.txt"
+)
+REACTOME_MAPPINGS_UNVERSIONED = (
+    "https://ftp.ebi.ac.uk/pub/databases/intact/current/various/reactome.dat"
+)
+
+
+# TODO it looks like it's probably also the case that
+#  this semantic space contains IDs for proteins/
+#  interactors. These need to be added too
+
+
+class IntactGetter(Obo):
+    """An ontology representation of Intact."""
+
+    bioversions_key = ontology = PREFIX
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return get_terms(force=force, version=self._version_or_raise)
+
+
+def get_terms(version: str, force: bool = False) -> Iterable[Term]:
+    """Get terms from the Contributor Roles Taxonomy via GitHub."""
+    cplx = _get_mappings(
+        COMPLEXPORTAL_MAPPINGS_UNVERSIONED, "complexportal", version=version, force=force
+    )
+    reactome = _get_mappings(
+        REACTOME_MAPPINGS_UNVERSIONED, "reactome", version=version, force=force
+    )
+    for intact_id in sorted(set(cplx).union(reactome)):
+        term = Term.from_triple(PREFIX, intact_id)
+        for complexportal_xref in sorted(cplx.get(intact_id, [])):
+            term.append_exact_match(complexportal_xref)
+        for reactome_xref in sorted(reactome.get(intact_id, [])):
+            term.append_xref(reactome_xref)
+        yield term
+
+
+def _get_mappings(
+    url: str, target_prefix: str, version: str, *, force: bool = False
+) -> dict[str, set[Reference]]:
+    path = ensure_path(PREFIX, url=url, version=version, force=force)
+    df = pd.read_csv(path, sep="\t", header=None, usecols=[0, 1])
+
+    dd = defaultdict(set)
+    for intact_id, target_identifier in df.values:
+        try:
+            obj = Reference(prefix=target_prefix, identifier=target_identifier)
+        except ValidationError:
+            tqdm.write(f"[intact:{intact_id}] invalid xref: {target_prefix}:{target_identifier}")
+            continue
+        dd[intact_id].add(obj)
+
+    return dict(dd)
+
+
+if __name__ == "__main__":
+    IntactGetter.cli()

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/struct/__init__.py

@@ -18,7 +18,7 @@ from .struct import (
     TypeDef,
     make_ad_hoc_ontology,
 )
-from .struct_utils import Annotation, Stanza
+from .struct_utils import Annotation, Stanza, StanzaType
 from .typedef import (
     derives_from,
     enables,
@@ -51,6 +51,7 @@ __all__ = [
     "Reference",
     "Referenced",
     "Stanza",
+    "StanzaType",
     "Synonym",
     "SynonymTypeDef",
     "Term",

{pyobo-0.12.1 → pyobo-0.12.3}/src/pyobo/struct/functional/ontology.py

@@ -250,7 +250,7 @@ class Import(Box):
 
 def get_rdf_graph_oracle(boxes: list[Box], *, prefix_map: dict[str, str]) -> Graph:
     """Serialize to turtle via OFN and conversion with ROBOT."""
-    from bioontologies.robot import convert
+    import bioontologies.robot
 
     ontology = Ontology(
         iri=EXAMPLE_ONTOLOGY_IRI,
@@ -265,7 +265,7 @@ def get_rdf_graph_oracle(boxes: list[Box], *, prefix_map: dict[str, str]) -> Gra
         ofn_path.write_text(text)
         ttl_path = stub.with_suffix(".ttl")
         try:
-            convert(ofn_path, ttl_path)
+            bioontologies.robot.convert(ofn_path, ttl_path)
         except subprocess.CalledProcessError:
             raise RuntimeError(f"failed to convert axioms from:\n\n{text}") from None
         graph.parse(ttl_path)

pyobo-0.12.3/src/pyobo/struct/obo/__init__.py

@@ -0,0 +1,9 @@
+"""OBO Flat file format."""
+
+from .reader import from_obo_path, from_obonet, from_str
+
+__all__ = [
+    "from_obo_path",
+    "from_obonet",
+    "from_str",
+]

{pyobo-0.12.1/src/pyobo → pyobo-0.12.3/src/pyobo/struct/obo}/reader.py

@@ -20,15 +20,6 @@ from curies.vocabulary import SynonymScope
 from more_itertools import pairwise
 from tqdm.auto import tqdm
 
-from .constants import DATE_FORMAT, PROVENANCE_PREFIXES
-from .identifier_utils import (
-    NotCURIEError,
-    ParseError,
-    UnparsableIRIError,
-    _is_valid_identifier,
-    _parse_str_or_curie_or_uri_helper,
-    get_rules,
-)
 from .reader_utils import (
     _chomp_axioms,
     _chomp_references,
@@ -36,7 +27,9 @@ from .reader_utils import (
     _chomp_typedef,
     _parse_provenance_list,
 )
-from .struct import (
+from .. import vocabulary as v
+from ..reference import OBOLiteral, _obo_parse_identifier
+from ..struct import (
     Obo,
     Reference,
     Synonym,
@@ -46,18 +39,26 @@ from .struct import (
     default_reference,
     make_ad_hoc_ontology,
 )
-from .struct import vocabulary as v
-from .struct.reference import OBOLiteral, _obo_parse_identifier
-from .struct.struct_utils import Annotation, Stanza
-from .struct.typedef import comment as has_comment
-from .struct.typedef import default_typedefs, has_ontology_root_term
-from .utils.cache import write_gzipped_graph
-from .utils.io import safe_open
-from .utils.misc import STATIC_VERSION_REWRITES, cleanup_version
+from ..struct_utils import Annotation, Stanza
+from ..typedef import comment as has_comment
+from ..typedef import default_typedefs, has_ontology_root_term
+from ...constants import DATE_FORMAT, PROVENANCE_PREFIXES
+from ...identifier_utils import (
+    NotCURIEError,
+    ParseError,
+    UnparsableIRIError,
+    _is_valid_identifier,
+    _parse_str_or_curie_or_uri_helper,
+    get_rules,
+)
+from ...utils.cache import write_gzipped_graph
+from ...utils.io import safe_open
+from ...utils.misc import STATIC_VERSION_REWRITES, cleanup_version
 
 __all__ = [
     "from_obo_path",
     "from_obonet",
+    "from_str",
 ]
 
 logger = logging.getLogger(__name__)
@@ -1353,6 +1354,8 @@ def _handle_prop(
             case None:
                 return None
 
+    return None
+
 
 def _get_prop(
     property_id: str,

pyobo-0.12.3/src/pyobo/struct/obograph/__init__.py

@@ -0,0 +1,16 @@
+"""I/O for OBO Graph JSON."""
+
+from .export import to_obograph, to_parsed_obograph, to_parsed_obograph_oracle, write_obograph
+from .reader import from_obograph, from_standardized_graph, read_obograph
+from .utils import assert_graph_equal
+
+__all__ = [
+    "assert_graph_equal",
+    "from_obograph",
+    "from_standardized_graph",
+    "read_obograph",
+    "to_obograph",
+    "to_parsed_obograph",
+    "to_parsed_obograph_oracle",
+    "write_obograph",
+]