pyobo 0.11.2__py3-none-any.whl → 0.12.1__py3-none-any.whl

pyobo/sources/pharmgkb/__init__.py

@@ -0,0 +1,15 @@
+"""Sources for PharmGKB."""
+
+from .pharmgkb_chemical import PharmGKBChemicalGetter
+from .pharmgkb_disease import PharmGKBDiseaseGetter
+from .pharmgkb_gene import PharmGKBGeneGetter
+from .pharmgkb_pathway import PharmGKBPathwayGetter
+from .pharmgkb_variant import PharmGKBVariantGetter
+
+__all__ = [
+    "PharmGKBChemicalGetter",
+    "PharmGKBDiseaseGetter",
+    "PharmGKBGeneGetter",
+    "PharmGKBPathwayGetter",
+    "PharmGKBVariantGetter",
+]

pyobo/sources/pharmgkb/pharmgkb_chemical.py

@@ -0,0 +1,89 @@
+"""An ontology representation of PharmGKB chemicals."""
+
+from collections.abc import Iterable
+
+import pandas as pd
+from tqdm import tqdm
+
+from pyobo import Obo, Reference, Term, default_reference
+from pyobo.sources.pharmgkb.utils import download_pharmgkb_tsv, split
+from pyobo.struct.typedef import has_inchi, has_smiles
+
+__all__ = [
+    "PharmGKBChemicalGetter",
+]
+
+PREFIX = "pharmgkb.drug"
+URL = "https://api.pharmgkb.org/v1/download/file/data/chemicals.zip"
+
+
+class PharmGKBChemicalGetter(Obo):
+    """An ontology representation of PharmGKB chemicals."""
+
+    ontology = bioversions_key = PREFIX
+    dynamic_version = True
+    typedefs = [has_inchi, has_smiles]
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return iter_terms(force=force)
+
+
+SKIP_PREFIXES = {"smiles", "inchi", "atc", "rxnorm", "pubchem.compound"}
+
+
+def iter_terms(force: bool = False) -> Iterable[Term]:
+    """Iterate over terms."""
+    df = download_pharmgkb_tsv(PREFIX, url=URL, inner="chemicals.tsv", force=force)
+
+    type_to_ref = {
+        typ: default_reference(PREFIX, typ.lower().replace(" ", "-").replace(",", ""), name=typ)
+        for typ in df["Type"].unique()
+    }
+    for x in type_to_ref.values():
+        yield Term(reference=x)
+
+    for _, row in df.iterrows():
+        term = Term.from_triple(PREFIX, identifier=row["PharmGKB Accession Id"], name=row["Name"])
+        term.append_parent(type_to_ref[row["Type"]])
+        if pd.notna(row["SMILES"]):
+            term.annotate_string(has_smiles, row["SMILES"])
+        if pd.notna(row["InChI"]):
+            term.annotate_string(has_inchi, row["InChI"])
+        for atc_id in split(row, "ATC Identifiers"):
+            term.append_exact_match(Reference(prefix="atc", identifier=atc_id))
+        for rxnorm_id in split(row, "RxNorm Identifiers"):
+            if len(rxnorm_id) > 7:
+                tqdm.write(f"invalid rxnorm luid (too long) - {rxnorm_id}")
+            else:
+                term.append_exact_match(Reference(prefix="rxnorm", identifier=rxnorm_id))
+        for pubchem_id in split(row, "PubChem Compound Identifiers"):
+            term.append_exact_match(Reference(prefix="pubchem.compound", identifier=pubchem_id))
+        for xref_curie in split(row, "External Vocabulary"):
+            try:
+                reference = Reference.from_curie(xref_curie)
+            except ValueError:
+                pass
+            else:
+                if reference.prefix not in SKIP_PREFIXES:
+                    term.append_exact_match(reference)
+        for xref_curie in split(row, "Cross-references"):
+            try:
+                reference = Reference.from_curie(xref_curie)
+            except ValueError:
+                pass
+            else:
+                if reference.prefix not in SKIP_PREFIXES:
+                    term.append_exact_match(reference)
+
+        for trade_name in split(row, "Trade names"):
+            # TODO use OMO term for trade name
+            term.append_synonym(trade_name)
+
+        # TODO add more
+
+        yield term
+
+
+if __name__ == "__main__":
+    PharmGKBChemicalGetter.cli()

pyobo/sources/pharmgkb/pharmgkb_disease.py

@@ -0,0 +1,77 @@
+"""An ontology representation of PharmGKB phenotypes."""
+
+from collections.abc import Iterable
+from typing import cast
+
+import pandas as pd
+
+from pyobo import Obo, Reference, Term
+from pyobo.sources.pharmgkb.utils import download_pharmgkb_tsv, parse_xrefs, split
+
+__all__ = [
+    "PharmGKBDiseaseGetter",
+]
+
+PREFIX = "pharmgkb.disease"
+URL = "https://api.pharmgkb.org/v1/download/file/data/phenotypes.zip"
+
+
+class PharmGKBDiseaseGetter(Obo):
+    """An ontology representation of PharmGKB phenotypes."""
+
+    ontology = bioversions_key = PREFIX
+    dynamic_version = True
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return iter_terms(force=force)
+
+
+def iter_terms(force: bool = False) -> Iterable[Term]:
+    """Iterate over terms.
+
+    :param force: Should the data be re-downloaded
+
+    :yields: Terms
+
+    1. PharmGKB Accession Id = Identifier assigned to this phenotype by PharmGKB
+    2. Name = Name PharmGKB uses for this phenotype
+    3. Alternate Names = Other known names for this phenotype, comma-separated
+    4. Cross-references = References to other resources in the form "resource:id",
+       comma-separated
+    5. External Vocabulary = Term for this phenotype in another vocabulary in the form
+       "vocabulary:id", comma-separated
+    """
+    df = download_pharmgkb_tsv(PREFIX, url=URL, inner="phenotypes.tsv", force=force)
+    for _, row in df.iterrows():
+        identifier = row["PharmGKB Accession Id"]
+        if pd.isna(identifier):
+            continue
+        name = row["Name"]
+        term = Term.from_triple(PREFIX, identifier=str(identifier), name=name)
+
+        synonyms = set()
+        for synonym in split(row, "Alternate Names"):
+            synonym = synonym.strip()
+            if synonym.casefold() == name.casefold():
+                continue
+            synonyms.add(synonym.strip('"'))
+        for synonym in sorted(synonyms):
+            term.append_synonym(synonym)
+        for xref in parse_xrefs(term, row):
+            term.append_xref(xref)
+
+        for xref_line in split(row, "External Vocabulary"):
+            xref_curie, _, _ = xref_line.strip('"').partition("(")
+            try:
+                xref = cast(Reference, Reference.from_curie(xref_curie))
+            except Exception:  # noqa:S110
+                pass  # this happens when there's a comma in the name, but not a problem
+            else:
+                term.append_xref(xref)
+
+        yield term
+
+
+if __name__ == "__main__":
+    PharmGKBDiseaseGetter.cli()

pyobo/sources/pharmgkb/pharmgkb_gene.py

@@ -0,0 +1,108 @@
+"""An ontology representation of PharmGKB genes."""
+
+from collections.abc import Iterable
+
+import pandas as pd
+
+from pyobo import Obo, Reference, Term
+from pyobo.sources.pharmgkb.utils import download_pharmgkb_tsv, parse_xrefs, split
+
+__all__ = [
+    "PharmGKBGeneGetter",
+]
+
+PREFIX = "pharmgkb.gene"
+URL = "https://api.pharmgkb.org/v1/download/file/data/genes.zip"
+
+
+class PharmGKBGeneGetter(Obo):
+    """An ontology representation of PharmGKB genes."""
+
+    ontology = bioversions_key = PREFIX
+    dynamic_version = True
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return iter_terms(force=force)
+
+
+def iter_terms(force: bool = False) -> Iterable[Term]:
+    """Iterate over terms.
+
+    :param force: Should the data be re-downloaded
+
+    :yields: Terms
+
+    1. PharmGKB Accession Id = Identifier assigned to this gene by PharmGKB
+    2. NCBI Gene ID = Identifier assigned to this gene by NCBI
+    3. HGNC ID = Identifier assigned to this gene by HGNC
+    4. Ensembl Id = Identifier assigned to this gene by Ensembl
+    5. Name = Canonical name for this gene (by HGNC)
+    6. Symbol = Canonical name for this gene (by HGNC)
+    7. Alternate Names = Other known names for this gene, comma-separated
+    8. Alternate Symbols = Other known symbols for this gene, comma-separated
+    9. Is VIP = "Yes" if PharmGKB has written a VIP annotation for this gene, "No"
+       otherwise
+    10. Has Variant Annotation = "Yes" if PharmGKB has written at least one variant
+        annotation for this gene, "No" otherwise
+    11. Cross-references = References to other resources in the form "resource:id",
+        comma-separated
+    12. Has CPIC Dosing Guideline = "Yes" if PharmGKB has annotated a CPIC guideline for
+        this gene, "No" otherwise
+    13. Chromosome = The chromosome this gene is on, in the form "chr##"
+    14. Chromosomal Start - GRCh37 = Where this gene starts on the chromosomal sequence
+        for NCBI GRCh37
+    15. Chromosomal Stop - GRCh37 = Where this gene stops on the chromosomal sequence
+        for NCBI GRCh37
+    16. Chromosomal Start - GRCh38 = Where this gene starts on the chromosomal sequence
+        for NCBI GRCh38
+    17. Chromosomal Stop - GRCh38 = Where this gene stops on the chromosomal sequence
+        for NCBI GRCh38
+    """
+    df = download_pharmgkb_tsv(PREFIX, url=URL, inner="genes.tsv", force=force)
+
+    skip_xrefs = {"ncbigene", "hgnc", "ensembl", "GeneCard"}
+    for _, row in df.iterrows():
+        identifier = row["PharmGKB Accession Id"]
+        if pd.isna(identifier):
+            continue
+
+        term = Term.from_triple(PREFIX, identifier=str(identifier), name=row["Name"])
+
+        ncbigene_ids = list(split(row, "NCBI Gene ID"))
+        if len(ncbigene_ids) == 1:
+            term.append_exact_match(Reference(prefix="ncbigene", identifier=ncbigene_ids[0]))
+        else:
+            for ncbigene_id in ncbigene_ids:
+                term.append_xref(Reference(prefix="ncbigene", identifier=ncbigene_id))
+
+        hgnc_ids = list(split(row, "HGNC ID"))
+        if len(hgnc_ids) == 1:
+            term.append_exact_match(Reference(prefix="hgnc", identifier=hgnc_ids[0]))
+        else:
+            for hgnc_id in hgnc_ids:
+                term.append_xref(Reference(prefix="hgnc", identifier=hgnc_id))
+
+        for ensembl_id in split(row, "Ensembl Id"):
+            term.append_xref(Reference(prefix="ensembl", identifier=ensembl_id))
+
+        for synonym in split(row, "Alternate Names"):
+            synonym = synonym.strip('"')
+            term.append_synonym(synonym)
+
+        # TODO symbol synonym type
+        if pd.notna(row["Symbol"]):
+            term.append_synonym(row["Symbol"])
+        for synonym in split(row, "Alternate Symbols"):
+            term.append_synonym(synonym)
+
+        for xref in parse_xrefs(term, row):
+            if xref.prefix in skip_xrefs:
+                continue
+            term.append_xref(xref)
+
+        yield term
+
+
+if __name__ == "__main__":
+    PharmGKBGeneGetter.cli()

pyobo/sources/pharmgkb/pharmgkb_pathway.py

@@ -0,0 +1,63 @@
+"""An ontology representation of PharmGKB pathways."""
+
+import zipfile
+from collections.abc import Iterable
+
+from pyobo import Obo, Term
+from pyobo.sources.pharmgkb.utils import download_pharmgkb
+
+__all__ = [
+    "PharmGKBPathwayGetter",
+]
+
+PREFIX = "pharmgkb.pathways"
+BIOPAX_URL = "https://api.pharmgkb.org/v1/download/file/data/pathways-biopax.zip"
+EXTENSION = ".owl"
+
+
+class PharmGKBPathwayGetter(Obo):
+    """An ontology representation of PharmGKB pathways."""
+
+    ontology = bioversions_key = PREFIX
+    dynamic_version = True
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return iter_terms(force=force)
+
+
+def iter_terms(force: bool = False) -> Iterable[Term]:
+    """Iterate over terms.
+
+    :param force: Should the data be re-downloaded
+
+    :yields: Terms
+
+    1. PharmGKB Accession Id = Identifier assigned to this phenotype by PharmGKB
+    2. Name = Name PharmGKB uses for this phenotype
+    3. Alternate Names = Other known names for this phenotype, comma-separated
+    4. Cross-references = References to other resources in the form "resource:id",
+       comma-separated
+    5. External Vocabulary = Term for this phenotype in another vocabulary in the form
+       "vocabulary:id", comma-separated
+    """
+    path = download_pharmgkb(PREFIX, url=BIOPAX_URL, force=force)
+    with zipfile.ZipFile(path) as zf:
+        for zip_info in zf.filelist:
+            if not zip_info.filename.endswith(EXTENSION):
+                continue
+            with zf.open(zip_info) as file:
+                yield _process_biopax(zip_info, file)
+
+
+def _process_biopax(path: zipfile.ZipInfo, file) -> Term:
+    fname = path.filename.removesuffix(EXTENSION).strip().replace("\r\n", " ")
+    identifier, _, name = fname.partition("-")
+    name = name.replace("_", " ")
+    term = Term.from_triple(PREFIX, identifier, name)
+    # TODO parse file with pybiopax to include members and provenance
+    return term
+
+
+if __name__ == "__main__":
+    PharmGKBPathwayGetter.cli()

pyobo/sources/pharmgkb/pharmgkb_variant.py

@@ -0,0 +1,84 @@
+"""An ontology representation of PharmGKB variants."""
+
+from collections.abc import Iterable
+
+import pandas as pd
+
+from pyobo import Obo, Reference, Term, TypeDef
+from pyobo.sources.pharmgkb.utils import download_pharmgkb_tsv, split
+
+__all__ = [
+    "PharmGKBVariantGetter",
+]
+
+PREFIX = "pharmgkb.variant"
+URL = "https://api.pharmgkb.org/v1/download/file/data/variants.zip"
+
+
+HAS_GENE_ASSOCIATION = TypeDef.default(
+    PREFIX, "hasGeneAssociation", name="has gene association", is_metadata_tag=True
+)
+
+
+class PharmGKBVariantGetter(Obo):
+    """An ontology representation of PharmGKB variants."""
+
+    ontology = bioversions_key = PREFIX
+    typedefs = [HAS_GENE_ASSOCIATION]
+    dynamic_version = True
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return iter_terms(force=force)
+
+
+def iter_terms(force: bool = False) -> Iterable[Term]:
+    """Iterate over terms.
+
+    :param force: Should the data be re-downloaded
+
+    :yields: Terms
+
+    1. Variant ID = The PharmGKB identifier for this variant
+    2. Variant Name = The PharmGKB name for this variant
+    3. Gene IDs = The PharmGKB identifiers for genes associated with this variant
+    4. Gene Symbols = The HGNC symbols for genes associated with this variant
+    5. Location = The location of this variation on a reference sequence (either RefSeq
+       or GenBank), if available. HGVS format when applicable
+    6. Variant Annotation count = The count of Variant Annotations done on this variant
+    7. Clinical Annotation count = The count of all Clinical Annotations done on this
+       variant
+    8. Level 1/2 Clinical Annotation count = The count of Level 1 or Level 2 ("top")
+       Clinical Annotations done on this variant
+    9. Guideline Annotation count = The count of Dosing Guideline Annotations of which
+       this variant is a part
+    10. Label Annotation count = The count of Drug Label Annotations in which this
+        variant is mentioned
+    11. Synonym
+    """
+    df = download_pharmgkb_tsv(PREFIX, url=URL, inner="variants.tsv", force=force)
+
+    for _, row in df.iterrows():
+        identifier = row["Variant ID"]
+        if pd.isna(identifier):
+            continue
+
+        term = Term.from_triple(PREFIX, identifier=str(identifier))
+
+        dbsnp_id = row["Variant Name"]
+        if pd.notna(dbsnp_id):
+            term.append_exact_match(Reference(prefix="dbsnp", identifier=dbsnp_id))
+
+        for gene_id, gene_name in zip(
+            split(row, "Gene IDs"), split(row, "Gene Symbols"), strict=False
+        ):
+            gene_ref = Reference(prefix="pharmgkb.gene", identifier=gene_id, name=gene_name)
+            term.annotate_object(HAS_GENE_ASSOCIATION, gene_ref)
+
+        # TODO location, like NC_000003.12:183917980
+
+        yield term
+
+
+if __name__ == "__main__":
+    PharmGKBVariantGetter.cli()

pyobo/sources/pharmgkb/utils.py

@@ -0,0 +1,86 @@
+"""Utilities for PharmGKB."""
+
+from collections.abc import Iterable
+from pathlib import Path
+from typing import cast
+
+import pandas as pd
+from pystow.utils import read_zipfile_csv
+from tqdm import tqdm
+
+from pyobo import Reference
+from pyobo.utils.path import ensure_path
+
+__all__ = [
+    "download_pharmgkb_tsv",
+]
+
+AGENT = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36"
+
+
+def download_pharmgkb(prefix: str, url: str, *, force: bool) -> Path:
+    """Download a file from PharmGKB, spoofing the user agent."""
+    return ensure_path(
+        prefix,
+        url=url,
+        backend="requests",
+        download_kwargs={
+            "headers": {
+                # This is required otherwise we get booted
+                "User-Agent": AGENT,
+            },
+        },
+        force=force,
+    )
+
+
+def download_pharmgkb_tsv(prefix: str, url: str, inner: str, *, force: bool) -> pd.DataFrame:
+    """Download PharmGKB data."""
+    path = download_pharmgkb(prefix, url=url, force=force)
+    df = read_zipfile_csv(path, inner_path=inner, dtype=str)
+    return df
+
+
+def split(row, key: str) -> Iterable[str]:
+    """Split the data."""
+    values = row.get(key)
+    if pd.isna(values) or not values:
+        return
+    try:
+        for value in values.split(","):
+            yield value.strip()
+    except AttributeError:
+        pass
+
+
+_MISSING_PREFIXES: set[str] = set()
+REPLACES = {
+    "URL:http://www.ncbi.nlm.nih.gov/omim/": "omim:",
+    "Comparative Toxicogenomics Database:": "mesh:",
+    "ModBase:": "uniprot:",
+    "RefSeq DNA:": "refseq:",
+    "RefSeq RNA:": "refseq:",
+    "RefSeq Protein:": "refseq:",
+    "UCSC Genome Browser:": "refseq:",
+}
+
+
+def parse_xrefs(term, row, key="Cross-references") -> Iterable[Reference]:
+    """Parse the cross-references."""
+    for xref_curie in split(row, key):
+        # HOXD@ is a valid genatlas identifier, see http://genatlas.medecine.univ-paris5.fr/fiche.php?symbol=HOXD@
+        # but this is broken, so skip them for now
+        if xref_curie.endswith("@"):
+            continue
+        for k, v in REPLACES.items():
+            if xref_curie.startswith(k):
+                xref_curie = xref_curie.replace(k, v)
+        try:
+            xref = cast(Reference, Reference.from_curie(xref_curie))
+        except ValueError:
+            p, _, _ = xref_curie.partition(":")
+            if p not in _MISSING_PREFIXES:
+                tqdm.write(f"[{term.curie}] could not parse xref: {xref_curie}")
+            _MISSING_PREFIXES.add(p)
+        else:
+            yield xref

pyobo/sources/pid.py

@@ -38,11 +38,6 @@ class PIDGetter(Obo):
         return iter_terms()
 
 
-def get_obo() -> Obo:
-    """Get NCI PID as OBO."""
-    return PIDGetter()
-
-
 def iter_networks(use_tqdm: bool = False, force: bool = False) -> Iterable[tuple[str, CX]]:
     """Iterate over NCI PID networks."""
     yield from ensure_ndex_network_set(
@@ -93,7 +88,7 @@ def iter_terms(force: bool = False) -> Iterable[Term]:
                 logger.debug(f"unmapped: {name}, {reference}")
 
         for hgnc_id, hgnc_symbol in genes:
-            term.append_relationship(
+            term.annotate_object(
                 has_participant, Reference(prefix="hgnc", identifier=hgnc_id, name=hgnc_symbol)
             )
 

pyobo/sources/pombase.py

@@ -8,7 +8,7 @@ import pandas as pd
 from tqdm.auto import tqdm
 
 import pyobo
-from pyobo import Reference
+from pyobo import Reference, TypeDef
 from pyobo.resources.so import get_so_name
 from pyobo.struct import Obo, Term, from_species, has_gene_product, orthologous
 from pyobo.utils.path import ensure_df
@@ -22,24 +22,20 @@ logger = logging.getLogger(__name__)
 PREFIX = "pombase"
 GENE_NAMES_URL = "https://www.pombase.org/data/names_and_identifiers/gene_IDs_names_products.tsv"
 ORTHOLOGS_URL = "https://www.pombase.org/data/orthologs/human-orthologs.txt.gz"
+CHROMOSOME = TypeDef.default(PREFIX, "chromosome", is_metadata_tag=True)
 
 
 class PomBaseGetter(Obo):
     """An ontology representation of PomBase's fission yeast gene nomenclature."""
 
     ontology = bioversions_key = PREFIX
-    typedefs = [from_species, has_gene_product, orthologous]
+    typedefs = [from_species, has_gene_product, orthologous, CHROMOSOME]
 
     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_obo(force: bool = False) -> Obo:
-    """Get OBO."""
-    return PomBaseGetter(force=force)
-
-
 #: A mapping from PomBase gene type to sequence ontology terms
 POMBASE_TO_SO = {
     # None: "0000704",  # gene,
@@ -89,13 +85,13 @@ def get_terms(version: str, force: bool = False) -> Iterable[Term]:
             name=symbol if pd.notna(symbol) else None,
             definition=name if pd.notna(name) else None,
         )
-        term.append_property("chromosome", chromosome[len("chromosome_") :])
+        term.annotate_string(CHROMOSOME, chromosome[len("chromosome_") :])
         term.append_parent(so[gtype])
         term.set_species(identifier="4896", name="Schizosaccharomyces pombe")
         for hgnc_id in identifier_to_hgnc_ids.get(identifier, []):
-            term.append_relationship(orthologous, Reference(prefix="hgnc", identifier=hgnc_id))
+            term.annotate_object(orthologous, Reference(prefix="hgnc", identifier=hgnc_id))
         if uniprot_id and pd.notna(uniprot_id):
-            term.append_relationship(
+            term.annotate_object(
                 has_gene_product, Reference(prefix="uniprot", identifier=uniprot_id)
             )
         if synonyms and pd.notna(synonyms):

pyobo/sources/pubchem.py

@@ -2,7 +2,7 @@
 
 import logging
 from collections.abc import Iterable, Mapping
-from typing import Optional
+from pathlib import Path
 
 import pandas as pd
 from bioregistry.utils import removeprefix
@@ -23,7 +23,7 @@ logger = logging.getLogger(__name__)
 PREFIX = "pubchem.compound"
 
 
-def _get_pubchem_extras_url(version: Optional[str], end: str) -> str:
+def _get_pubchem_extras_url(version: str | None, end: str) -> str:
     if version is None:
         version = get_version("pubchem")
     return f"ftp://ftp.ncbi.nlm.nih.gov/pubchem/Compound/Monthly/{version}/Extras/{end}"
@@ -40,11 +40,6 @@ class PubChemCompoundGetter(Obo):
         return get_terms(version=self._version_or_raise, force=force)
 
 
-def get_obo(force: bool = False) -> Obo:
-    """Get PubChem Compound OBO."""
-    return PubChemCompoundGetter(force=force)
-
-
 def _get_cid_smiles_df(version: str) -> pd.DataFrame:
     url = _get_pubchem_extras_url(version, "CID-SMILES.gz")
     return ensure_df(PREFIX, url=url, version=version, dtype=str)
@@ -97,7 +92,7 @@ def get_pubchem_id_to_mesh_id(version: str) -> Mapping[str, str]:
     return dict(df.values)
 
 
-def _ensure_cid_name_path(*, version: Optional[str] = None, force: bool = False) -> str:
+def _ensure_cid_name_path(*, version: str | None = None, force: bool = False) -> Path:
     if version is None:
         version = get_version("pubchem")
     # 2 tab-separated columns: compound_id, name
@@ -145,4 +140,4 @@ def get_terms(*, version: str, use_tqdm: bool = True, force: bool = False) -> It
 
 
 if __name__ == "__main__":
-    get_obo().write_default()
+    PubChemCompoundGetter.cli()