pyobo 0.12.15__tar.gz → 0.12.16__tar.gz

{pyobo-0.12.15 → pyobo-0.12.16}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyobo
-Version: 0.12.15
+Version: 0.12.16
 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
@@ -36,9 +36,8 @@ Requires-Dist: humanize
 Requires-Dist: tabulate
 Requires-Dist: cachier
 Requires-Dist: pystow>=0.7.5
-Requires-Dist: bioversions>=0.8.101
+Requires-Dist: bioversions>=0.8.243
 Requires-Dist: bioregistry>=0.12.30
-Requires-Dist: bioontologies>=0.7.2
 Requires-Dist: ssslm>=0.0.13
 Requires-Dist: zenodo-client>=0.4.1
 Requires-Dist: class-resolver>=0.6.0
@@ -52,10 +51,14 @@ Requires-Dist: chembl-downloader
 Requires-Dist: umls-downloader>=0.1.3
 Requires-Dist: clinicaltrials-downloader>=0.0.2
 Requires-Dist: nih-reporter-downloader>=0.0.1
+Requires-Dist: ror-downloader>=0.0.4
 Requires-Dist: typing-extensions
 Requires-Dist: rdflib
 Requires-Dist: obographs>=0.0.8
-Requires-Dist: sssom-pydantic>=0.1.1
+Requires-Dist: sssom-pydantic>=0.2.0
+Requires-Dist: pytz
+Requires-Dist: robot-obo-tool
+Requires-Dist: pyperclip
 Requires-Dist: psycopg2-binary ; extra == 'drugcentral'
 Requires-Dist: ssslm[gilda] ; extra == 'gilda'
 Requires-Dist: ssslm[gilda-slim] ; extra == 'gilda-slim'
@@ -72,10 +75,10 @@ Maintainer: Charles Tapley Hoyt
 Maintainer-email: Charles Tapley Hoyt <cthoyt@gmail.com>
 Requires-Python: >=3.10
 Project-URL: Bug Tracker, https://github.com/biopragmatics/pyobo/issues
-Project-URL: Documentation, https://pyobo.readthedocs.io
-Project-URL: Funding, https://github.com/sponsors/cthoyt
 Project-URL: Homepage, https://github.com/biopragmatics/pyobo
 Project-URL: Repository, https://github.com/biopragmatics/pyobo.git
+Project-URL: Documentation, https://pyobo.readthedocs.io
+Project-URL: Funding, https://github.com/sponsors/cthoyt
 Provides-Extra: drugcentral
 Provides-Extra: gilda
 Provides-Extra: gilda-slim

{pyobo-0.12.15 → pyobo-0.12.16}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "pyobo"
-version = "0.12.15"
+version = "0.12.16"
 description = "A python package for handling and generating OBO"
 readme = "README.md"
 authors = [
@@ -69,9 +69,8 @@ dependencies = [
     "tabulate",
     "cachier",
     "pystow>=0.7.5",
-    "bioversions>=0.8.101",
+    "bioversions>=0.8.243",
     "bioregistry>=0.12.30",
-    "bioontologies>=0.7.2",
     "ssslm>=0.0.13",
     "zenodo-client>=0.4.1",
     "class_resolver>=0.6.0",
@@ -86,10 +85,14 @@ dependencies = [
     "umls_downloader>=0.1.3",
     "clinicaltrials_downloader>=0.0.2",
     "nih-reporter-downloader>=0.0.1",
+    "ror-downloader>=0.0.4",
     "typing_extensions",
     "rdflib",
     "obographs>=0.0.8",
-    "sssom_pydantic>=0.1.1",
+    "sssom_pydantic>=0.2.0",
+    "pytz",
+    "robot-obo-tool",
+    "pyperclip",
 ]
 
 # see https://peps.python.org/pep-0735/ and https://docs.astral.sh/uv/concepts/dependencies/#dependency-groups
@@ -99,11 +102,11 @@ tests = [
     "coverage[toml]",
 ]
 docs = [
-    "sphinx>=8",
+    "sphinx>=8,<9",
     "sphinx-rtd-theme>=3.0",
     "sphinx-click", 
     "sphinx_automodapi",
-    "sphinx_toolbox",
+    "sphinx_toolbox<=4.0.0",
 ]
 lint = [
     "ruff",
@@ -315,7 +318,7 @@ known-first-party = [
 docstring-code-format = true
 
 [tool.bumpversion]
-current_version = "0.12.15"
+current_version = "0.12.16"
 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-0.12.15 → pyobo-0.12.16}/src/pyobo/api/xrefs.py

@@ -124,7 +124,11 @@ def get_semantic_mappings(
     )
     if converter is None:
         converter = get_converter()
-    return [row_to_semantic_mapping(row, converter=converter) for _, row in df.iterrows()]
+    return [
+        # TODO upstream this into sssom-pydantic?
+        row_to_semantic_mapping({k: v for k, v in row.items() if pd.notna(v)}, converter=converter)
+        for _, row in df.iterrows()
+    ]
 
 
 def get_mappings_df(

{pyobo-0.12.15 → pyobo-0.12.16}/src/pyobo/cli/cli.py

@@ -10,6 +10,7 @@ import click
 
 from .database import main as database_main
 from .lookup import lookup
+from .obo_lexical_review import obo_lexical_review
 from ..constants import GLOBAL_SKIP, RAW_DIRECTORY
 from ..plugins import has_nomenclature_plugin
 
@@ -101,6 +102,7 @@ def _no_download() -> set[str]:
 
 main.add_command(lookup)
 main.add_command(database_main)
+main.add_command(obo_lexical_review)
 
 if __name__ == "__main__":
     main()

{pyobo-0.12.15 → pyobo-0.12.16}/src/pyobo/cli/lookup.py

@@ -1,8 +1,12 @@
 """CLI for PyOBO lookups."""
 
+from __future__ import annotations
+
+import inspect
 import json
 import sys
-from collections.abc import Mapping
+from collections.abc import Iterable, Mapping
+from typing import TYPE_CHECKING
 
 import click
 from more_click import verbose_option
@@ -17,7 +21,10 @@ from .utils import (
     strict_option,
     version_option,
 )
-from ..constants import LookupKwargs
+from ..constants import GetOntologyKwargs, LookupKwargs
+
+if TYPE_CHECKING:
+    from curies import Reference
 
 __all__ = [
     "lookup",
@@ -31,9 +38,12 @@ def lookup():
 
 def lookup_annotate(f: Clickable) -> Clickable:
     """Add appropriate decorators to lookup CLI functions."""
+    signature = inspect.signature(f)
+    param = signature.parameters["kwargs"]
+    if param.kind is not inspect.Parameter.VAR_KEYWORD:
+        raise ValueError("programmer error")
     for decorator in [
         lookup.command(),
-        prefix_argument,
         verbose_option,
         force_option,
         force_process_option,
@@ -41,6 +51,15 @@ def lookup_annotate(f: Clickable) -> Clickable:
         version_option,
     ]:
         f = decorator(f)
+
+    if param.annotation is None:
+        pass
+    if param.annotation in {"Unpack[LookupKwargs]", Unpack[LookupKwargs]}:
+        f = prefix_argument(f)
+    elif param.annotation in {"Unpack[GetOntologyKwargs]", Unpack[GetOntologyKwargs]}:
+        pass
+    else:
+        raise ValueError(f"unknown parameter type for {f}: {param.annotation}")
     return f
 
 
@@ -249,33 +268,35 @@ def hierarchy(
 
 
 @lookup_annotate
-@click.argument("identifier")
-def ancestors(
-    identifier: str,
-    **kwargs: Unpack[LookupKwargs],
-) -> None:
+@click.argument("curie")
+def ancestors(curie: str, **kwargs: Unpack[GetOntologyKwargs]) -> None:
     """Look up ancestors."""
-    from ..api import get_ancestors, get_name
+    from ..api import get_ancestors
 
-    # note, prefix is passed via kwargs
-    ancestors = get_ancestors(identifier=identifier, **kwargs)
-    for ancestor in sorted(ancestors or []):
-        click.echo(f"{ancestor.curie}\t{get_name(ancestor, version=kwargs['version'])}")
+    ancestors = get_ancestors(curie, **kwargs)
+    _list_curies(ancestors, **kwargs)
 
 
 @lookup_annotate
-@click.argument("identifier")
-def descendants(
-    identifier: str,
-    **kwargs: Unpack[LookupKwargs],
-) -> None:
+@click.argument("curie")
+def descendants(curie: str, **kwargs: Unpack[GetOntologyKwargs]) -> None:
     """Look up descendants."""
-    from ..api import get_descendants, get_name
+    from ..api import get_descendants
+
+    descendants = get_descendants(curie, **kwargs)
+    _list_curies(descendants, **kwargs)
+
+
+def _list_curies(
+    references: Iterable[Reference] | None, **kwargs: Unpack[GetOntologyKwargs]
+) -> None:
+    if not references:
+        return
+
+    from ..api import get_name
 
-    # note, prefix is passed via kwargs
-    descendants = get_descendants(identifier=identifier, **kwargs)
-    for descendant in sorted(descendants or []):
-        click.echo(f"{descendant.curie}\t{get_name(descendant, version=kwargs['version'])}")
+    for reference in sorted(references or []):
+        click.echo(f"{reference.curie}\t{get_name(reference, version=kwargs['version'])}")
 
 
 @lookup_annotate

pyobo-0.12.16/src/pyobo/cli/obo_lexical_review.py

@@ -0,0 +1,207 @@
+# /// script
+# requires-python = ">=3.11"
+# dependencies = [
+#     "click>=8.3.1",
+#     "obographs>=0.0.8",
+#     "pyperclip>=1.11.0",
+#     "robot-obo-tool>=0.0.1",
+#     "ssslm[gilda-slim]>=0.1.3",
+#     "tabulate>=0.9.0",
+#     "tqdm>=4.67.3",
+# ]
+# ///
+
+"""Implement lexical review for an ontology."""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import click
+
+if TYPE_CHECKING:
+    import obographs
+    import ssslm
+
+INDEX_URL = "https://github.com/biopragmatics/biolexica/raw/main/lexica/obo/obo.ssslm.tsv.gz"
+UPPER = {"ncit"}
+
+
+@click.command()
+@click.argument("prefix")
+@click.option(
+    "--location",
+    help="Local path or URL to an OBO Graph JSON file or OWL file. If not given, will try and look up through the OBO PURL system",
+)
+@click.option(
+    "--uri-prefix",
+    help="Local path to an OBO Graph JSON file. If not given, will try and look up through the OBO PURL system",
+)
+@click.option("--index-url", default=INDEX_URL, show_default=True)
+@click.option("--show-passed", is_flag=True)
+@click.option("--skip-upper", is_flag=True, help=f"if true, skip upper level ontologies {UPPER}")
+@click.option("--index-force", is_flag=True, help="if true, force re-downloading the lexical index")
+def obo_lexical_review(
+    prefix: str,
+    location: str | None,
+    uri_prefix: str | None,
+    index_url: str,
+    show_passed: bool,
+    skip_upper: bool,
+    index_force: bool,
+) -> None:
+    """Make a lexical review of an ontology."""
+    import sys
+    import time
+
+    import pyperclip
+    import pystow
+    import ssslm
+    from tabulate import tabulate
+
+    module = pystow.module("pyobo", "obo-lexical-review")
+
+    args = " ".join(sys.argv[1:])
+    output = f"Analysis of {prefix} run on {time.asctime()} with the following command:\n\n```console\n$ uvx pyobo obo-lexical-review {args}\n```\n\n"
+
+    graph_document, uri_prefix = _get_graph_document(
+        prefix=prefix,
+        uri_prefix=uri_prefix,
+        ontology_path=location,
+    )
+
+    click.echo(f"Loading lexical index from {index_url} using SSSLM")
+    path = module.ensure(url=index_url, force=index_force)
+    grounder = ssslm.make_grounder(path)
+    click.echo("Done loading lexical index")
+
+    passed, failed = _get_calls(
+        graph_document=graph_document,
+        matcher=grounder,
+        uri_prefix=uri_prefix,
+        skip_upper=skip_upper,
+    )
+
+    total = len(passed) + len(failed)
+
+    if passed and show_passed:
+        passed_table = tabulate(passed, headers=["LUID", "Name"], tablefmt="github")
+        passed_msg = f"## Passed Nodes ({len(passed):,}/{total:,}; {len(passed) / total:.1%})\n\n{passed_table}\n\n"
+        output += passed_msg
+
+    if failed:
+        rows = []
+        for luid, name, matches in failed:
+            rows.append((luid, name, *_parts(matches[0])))
+            for match in matches[1:]:
+                rows.append(("", "", *_parts(match)))
+        failed_table = tabulate(
+            rows, headers=[prefix, "name", "obo-curie", "obo-name", "obo-score"], tablefmt="github"
+        )
+        failed_message = f"## Failed Nodes ({len(failed):,}/{total:,}; {len(failed) / total:.1%})\n\n{failed_table}\n\n"
+        output += failed_message
+
+    click.echo(output)
+    click.echo("Finished! automatically copied to the clipboard, e.g., for pasting into GitHub)")
+    pyperclip.copy(output)
+
+
+def _parts(match: ssslm.Match) -> tuple[str, str, float]:
+    return (
+        f"[`{match.curie}`](https://semantic.farm/{match.curie})",
+        match.name or "",
+        round(match.score, 3),
+    )
+
+
+def _get_graph_document(
+    prefix: str, uri_prefix: str | None = None, ontology_path: str | None = None
+) -> tuple[obographs.GraphDocument, str]:
+    from pathlib import Path
+
+    import obographs
+    import robot_obo_tool
+
+    if uri_prefix is None:
+        uri_prefix = f"http://purl.obolibrary.org/obo/{prefix}_"
+        click.echo(f"Inferred URI prefix from given OBO CURIE prefix: {uri_prefix}")
+    if ontology_path is None:
+        ontology_path = f"https://purl.obolibrary.org/obo/{prefix.lower()}.json"
+        click.echo(f"No ontology path given, guessing it's available at {ontology_path}")
+    if ontology_path.endswith(".json"):
+        click.echo(f"reading OBO Graph JSON from {ontology_path}")
+        graph_documents = obographs.read(ontology_path, squeeze=False, timeout=60)
+    else:
+        import tempfile
+
+        with tempfile.TemporaryDirectory() as tmpdir:
+            tmppath = Path(tmpdir).joinpath("temp.json")
+            click.echo(
+                "given ontology path does not end with JSON. implicitly converting to OBO Graph JSON using ROBOT"
+            )
+            robot_obo_tool.convert(
+                input_path=ontology_path,
+                output_path=tmppath,
+                check=False,
+                merge=False,
+                reason=False,
+            )
+            click.echo("reading converted OBO Graph JSON")
+            graph_documents = obographs.read(tmppath, squeeze=False)
+
+    return graph_documents, uri_prefix
+
+
+def _get_calls(
+    *,
+    graph_document: obographs.GraphDocument,
+    matcher: ssslm.Matcher,
+    uri_prefix: str,
+    skip_upper: bool = False,
+) -> tuple[list[tuple[str, str]], list[tuple[str, str, list[ssslm.Match]]]]:
+    """Get matches."""
+    from tqdm import tqdm
+
+    passed = []
+    failed = []
+    total = 0
+    skipped = 0
+    for graph in tqdm(graph_document.graphs, unit="graph"):
+        for node in tqdm(sorted(graph.nodes, key=lambda n: n.id), unit="node", leave=False):
+            if node.id is None:
+                continue
+
+            total += 1
+            if not node.id.startswith(uri_prefix):
+                skipped += 1
+                continue
+
+            if not node.lbl:
+                continue
+
+            local_unique_identifier = node.id[len(uri_prefix) :]
+
+            matches: list[ssslm.Match] = []
+            matches.extend(matcher.get_matches(node.lbl))
+            if node.meta is not None and node.meta.synonyms is not None:
+                matches.extend(
+                    match
+                    for synonym in node.meta.synonyms
+                    for match in matcher.get_matches(synonym.val)
+                )
+
+            # there are a lot of NCIT matches, which aren't that informative
+            # since OBO doesn't mind duplicating these terms
+            if skip_upper:
+                matches = [m for m in matches if m.prefix not in UPPER]
+
+            if not matches:
+                passed.append((local_unique_identifier, node.lbl))
+            else:
+                failed.append((local_unique_identifier, node.lbl, matches))
+
+    return passed, failed
+
+
+if __name__ == "__main__":
+    obo_lexical_review()

{pyobo-0.12.15 → pyobo-0.12.16}/src/pyobo/getters.py

@@ -17,7 +17,6 @@ from pathlib import Path
 from textwrap import indent
 from typing import Any, TypeVar
 
-import bioontologies.robot
 import bioregistry
 import click
 import pystow.utils
@@ -80,10 +79,10 @@ REQUIRES_NO_ROBOT_CHECK = {
 
 
 def _convert_to_obo(path: Path) -> Path:
-    import bioontologies.robot
+    import robot_obo_tool
 
     _converted_obo_path = path.with_suffix(".obo")
-    bioontologies.robot.convert(path, _converted_obo_path, check=False)
+    robot_obo_tool.convert(path, _converted_obo_path, check=False)
     return _converted_obo_path
 
 
@@ -338,6 +337,8 @@ def iter_helper_helper(
 
     :yields: A prefix and the result of the callable ``f``
     """
+    from robot_obo_tool import ROBOTError
+
     strict = kwargs.get("strict", True)
     prefixes = list(
         _prefixes(
@@ -382,7 +383,7 @@ def iter_helper_helper(
             if "DrugBank" not in str(e):
                 raise
             logger.warning("[drugbank] invalid credentials")
-        except (subprocess.CalledProcessError, bioontologies.robot.ROBOTError):
+        except (subprocess.CalledProcessError, ROBOTError):
             logger.warning("[%s] ROBOT was unable to convert OWL to OBO", prefix)
         except ValueError as e:
             if _is_xml(e):