pyobo 0.11.2__py3-none-any.whl → 0.12.0__py3-none-any.whl

pyobo/sources/ncbi/ncbi_gc.py

@@ -0,0 +1,162 @@
+"""Convert NCBI Genetic Codes to an ontology.
+
+.. seealso::
+
+    https://www.ncbi.nlm.nih.gov/Taxonomy/taxonomyhome.html/index.cgi?chapter=cgencodes
+"""
+
+from collections.abc import Iterable
+
+from pyobo import default_reference
+from pyobo.struct import CHARLIE_TERM, HUMAN_TERM, PYOBO_INJECTED, Obo, Reference, Term, TypeDef
+from pyobo.struct.typedef import comment, has_contributor, see_also, term_replaced_by
+from pyobo.utils.path import ensure_path
+
+PREFIX = "ncbi.gc"
+URI_PREFIX = (
+    "https://www.ncbi.nlm.nih.gov/Taxonomy/taxonomyhome.html/index.cgi?chapter=cgencodes#SG"
+)
+URL = "ftp://ftp.ncbi.nih.gov/entrez/misc/data/gc.prt"
+VERSION = "4.6"
+
+GC_ROOT = default_reference(prefix=PREFIX, identifier="root", name="genetic code translation table")
+NCBITAXON_ROOT = Reference(prefix="NCBITaxon", identifier="1", name="root")
+
+has_gc_code = TypeDef(
+    reference=default_reference(
+        prefix=PREFIX,
+        identifier="hasGeneticCodeTranslationTable",
+        name="has genetic code translation table",
+    ),
+    definition="Connects a taxonomy term to a genetic code translation table",
+    domain=NCBITAXON_ROOT,
+    range=GC_ROOT,
+).append_contributor(CHARLIE_TERM)
+
+NUCLEAR_GENETIC_CODE = default_reference(
+    prefix=PREFIX, identifier="nuclear-genetic-code", name="nuclear genetic code translation table"
+)
+MITOCHONDRIAL_GENETIC_CODE = default_reference(
+    prefix=PREFIX,
+    identifier="mitochondrial-genetic-code",
+    name="mitochondrial genetic code translation table",
+)
+PLASTID_GENETIC_CODE = default_reference(
+    prefix=PREFIX, identifier="plastid-genetic-code", name="plastid genetic code translation table"
+)
+NUCLEUS = Reference(prefix="GO", identifier="0005634", name="nucleus")
+MITOCHONDIA = Reference(prefix="GO", identifier="0005739", name="mitochondrion")
+PLASTID = Reference(prefix="GO", identifier="0009536", name="plastid")
+
+CATEGORY_TO_CELLULAR_COMPONENT = {
+    NUCLEAR_GENETIC_CODE: NUCLEUS,
+    MITOCHONDRIAL_GENETIC_CODE: MITOCHONDIA,
+    PLASTID_GENETIC_CODE: PLASTID,
+}
+CATEGORY_TO_TABLES = {
+    NUCLEAR_GENETIC_CODE: [12, 31, 6, 28, 10, 27, 29, 26, 30, 15],
+    MITOCHONDRIAL_GENETIC_CODE: [14, 13, 16, 9, 5, 4, 22, 23, 21, 2, 3, 24],
+    PLASTID_GENETIC_CODE: [11, 32],
+}
+TABLE_TO_CATEGORY = {
+    str(value): key for key, values in CATEGORY_TO_TABLES.items() for value in values
+}
+
+
+class NCBIGCGetter(Obo):
+    """Get terms in GC."""
+
+    ontology = PREFIX
+    static_version = VERSION
+    root_terms = [GC_ROOT]
+    typedefs = [has_gc_code, has_contributor, see_also, comment, term_replaced_by]
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        return get_terms()
+
+
+def get_terms() -> Iterable[Term]:
+    """Get terms for GC."""
+    yield CHARLIE_TERM
+    yield Term(reference=NCBITAXON_ROOT)
+    yield HUMAN_TERM
+
+    path = ensure_path(PREFIX, url=URL)
+    # first, remove comment lines
+    lines = [
+        line.strip()
+        for line in path.read_text().splitlines()
+        if not line.startswith("--") and line.strip()
+    ]
+
+    lines = lines[1:-2]
+    entries: list[dict[str, str]] = []
+    entry: dict[str, str] = {}
+    for line in lines:
+        # start a new entry
+        if line == "{":
+            if entry:
+                entries.append(entry)
+            entry = {}
+        elif line == "},":
+            pass
+        else:
+            key, data = line.split(" ", 1)
+            if key == "name":
+                data = data.lstrip('"')
+                if data.startswith("SGC"):
+                    key = "symbol"
+                entry[key] = data.rstrip(",").rstrip().rstrip('"')
+            elif key == "id":
+                entry["identifier"] = data.rstrip(",").rstrip()
+
+    yield (
+        Term(
+            reference=GC_ROOT,
+            definition="A table for translating codons into amino acids. This can change for "
+            "different taxa, or be different in different organelles that include genetic information.",
+        )
+        .append_contributor(CHARLIE_TERM)
+        .append_comment(PYOBO_INJECTED)
+    )
+
+    for reference in CATEGORY_TO_TABLES:
+        term = Term(reference=reference)
+        term.append_parent(GC_ROOT)
+        term.append_contributor(CHARLIE_TERM)
+        term.append_comment(PYOBO_INJECTED)
+        if substructure := CATEGORY_TO_CELLULAR_COMPONENT.get(reference):
+            term.append_see_also(substructure)
+        yield term
+
+    for entry in entries:
+        identifier = entry["identifier"]
+        term = Term.from_triple(PREFIX, identifier, entry["name"])
+        term.append_parent(TABLE_TO_CATEGORY.get(identifier, GC_ROOT))
+        # TODO if symbol is available, what does it mean?
+        yield term
+
+    yield (
+        Term(
+            reference=Reference(prefix=PREFIX, identifier="7"),
+            is_obsolete=True,
+        )
+        .append_replaced_by(Reference(prefix=PREFIX, identifier="4"))
+        .append_comment("Kinetoplast code now merged in code id 4, as of 1995.")
+    )
+    yield (
+        Term(
+            reference=Reference(prefix=PREFIX, identifier="8"),
+            is_obsolete=True,
+        )
+        .append_replaced_by(Reference(prefix=PREFIX, identifier="1"))
+        .append_comment("all plant chloroplast differences due to RNA edit, as of 1995.")
+    )
+
+    for cellular_component in CATEGORY_TO_CELLULAR_COMPONENT.values():
+        yield Term(reference=cellular_component)
+
+
+if __name__ == "__main__":
+    NCBIGCGetter.cli()

pyobo/sources/{ncbigene.py → ncbi/ncbigene.py}

@@ -7,8 +7,8 @@ import bioregistry
 import pandas as pd
 from tqdm.auto import tqdm
 
-from ..struct import Obo, Reference, Term, from_species
-from ..utils.path import ensure_df
+from ...struct import Obo, Reference, Term, from_species
+from ...utils.path import ensure_df
 
 __all__ = [
     "NCBIGeneGetter",
@@ -34,7 +34,7 @@ CONSORTIUM_SPECIES_MAPPING = {
 }
 
 GENE_INFO_URL = "ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/gene_info.gz"
-#: Columns fro gene_info.gz that are used
+#: Columns for gene_info.gz that are used
 GENE_INFO_COLUMNS = [
     "#tax_id",
     "GeneID",
@@ -93,11 +93,6 @@ class NCBIGeneGetter(Obo):
         return get_terms(force=force)
 
 
-def get_obo(force: bool = False) -> Obo:
-    """Get Entrez as OBO."""
-    return NCBIGeneGetter(force=force)
-
-
 def get_gene_info_df(force: bool = False) -> pd.DataFrame:
     """Get the gene info dataframe."""
     return ensure_df(
@@ -111,17 +106,16 @@ def get_gene_info_df(force: bool = False) -> pd.DataFrame:
     )
 
 
-"""xref_mapping was obtained from:
-
-namespaces = set()
-for xrefs in df[df['dbXrefs'].notna()]['dbXrefs']:
-    for xref in xrefs.split('|'):
-        namespaces.add(xref.split(':')[0])
+def _get_xref_mapping() -> list[str]:
+    namespaces: set[str] = set()
+    df = get_gene_info_df()
+    for xrefs in df[df["dbXrefs"].notna()]["dbXrefs"]:
+        for xref in xrefs.split("|"):
+            namespaces.add(xref.split(":")[0])
+    return sorted(namespaces, key=str.casefold)
 
-print('namespaces:')
-print(*sorted(namespaces), sep='\n')
-"""
 
+# this was retrieved from :func:`_get_xref_mapping`
 xref_mapping = {
     "APHIDBASE",
     "ASAP",
@@ -157,7 +151,12 @@ xref_mapping = {x.lower() for x in xref_mapping}
 
 
 def get_terms(force: bool = False) -> Iterable[Term]:
-    """Get Entrez terms."""
+    """Get Entrez terms.
+
+    :param force: should re-download be forced?
+
+    :yields: terms for each line
+    """
     df = get_gene_info_df(force=force)
 
     it = tqdm(
@@ -192,4 +191,4 @@ def get_terms(force: bool = False) -> Iterable[Term]:
 
 
 if __name__ == "__main__":
-    get_obo().write_default()
+    NCBIGeneGetter.cli()

pyobo/sources/nih_reporter.py

@@ -0,0 +1,60 @@
+"""A source for NIH RePORTER projects."""
+
+from collections.abc import Iterable
+
+import pandas as pd
+from nih_reporter_downloader import get_projects_df
+
+from pyobo import Reference
+from pyobo.struct import CHARLIE_TERM, HUMAN_TERM, PYOBO_INJECTED, Obo, Term, default_reference
+
+__all__ = [
+    "NIHReporterGetter",
+]
+
+PREFIX = "nihreporter.project"
+PROJECTS_SUBSET = [
+    "APPLICATION_ID",
+    "PROJECT_TITLE",
+]
+
+PROJECT_TERM = (
+    Term(reference=default_reference(PREFIX, "project", name="project"))
+    .append_contributor(CHARLIE_TERM)
+    .append_comment(PYOBO_INJECTED)
+)
+
+
+class NIHReporterGetter(Obo):
+    """An ontology representation of NIH RePORTER."""
+
+    ontology = PREFIX
+    dynamic_version = True
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over terms in the ontology."""
+        yield CHARLIE_TERM
+        yield HUMAN_TERM
+        yield PROJECT_TERM
+        yield from iterate_nih_reporter_projects()
+
+
+def iterate_nih_reporter_projects() -> Iterable[Term]:
+    """Iterate over NIH RePORTER projects."""
+    projects_df = get_projects_df()
+    for identifier, name in projects_df[PROJECTS_SUBSET].values:
+        term = Term(
+            reference=Reference(
+                prefix=PREFIX,
+                identifier=str(identifier),
+                name=name.replace("\n", " ") if pd.notna(name) else None,
+            ),
+            type="Instance",
+        )
+        term.append_parent(PROJECT_TERM)
+        # TODO there is a lot more information that can be added here
+        yield term
+
+
+if __name__ == "__main__":
+    NIHReporterGetter.cli()

pyobo/sources/nlm/__init__.py

@@ -0,0 +1,9 @@
+"""Sources from NLM."""
+
+from .nlm_catalog import NLMCatalogGetter
+from .nlm_publisher import NLMPublisherGetter
+
+__all__ = [
+    "NLMCatalogGetter",
+    "NLMPublisherGetter",
+]

pyobo/sources/nlm/nlm_catalog.py

@@ -0,0 +1,48 @@
+"""Converter for NLM Providers."""
+
+from collections.abc import Iterable
+
+from pyobo.sources.nlm.utils import (
+    JOURNAL_TERM,
+    PREFIX_CATALOG,
+    PUBLISHED_IN,
+    PUBLISHER_TERM,
+    get_journals,
+    get_publishers,
+)
+from pyobo.struct import CHARLIE_TERM, HUMAN_TERM, Obo, Term
+from pyobo.struct.typedef import exact_match, has_end_date, has_start_date
+
+__all__ = [
+    "NLMCatalogGetter",
+]
+
+
+class NLMCatalogGetter(Obo):
+    """An ontology representation of NLM Providers."""
+
+    bioversions_key = ontology = PREFIX_CATALOG
+    dynamic_version = True
+    typedefs = [PUBLISHED_IN, has_end_date, has_start_date, exact_match]
+    root_terms = [JOURNAL_TERM.reference, PUBLISHER_TERM.reference]
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over journal terms for NLM Catalog."""
+        yield from get_terms(force=force)
+
+
+def get_terms(*, force: bool = False) -> Iterable[Term]:
+    """Get NLM catalog terms."""
+    yield JOURNAL_TERM
+    yield PUBLISHER_TERM
+    yield CHARLIE_TERM
+    yield HUMAN_TERM
+
+    journal_id_to_publisher_key = get_publishers(force=force)
+    yield from sorted(set(journal_id_to_publisher_key.values()))
+
+    yield from get_journals(force=force, journal_id_to_publisher_key=journal_id_to_publisher_key)
+
+
+if __name__ == "__main__":
+    NLMCatalogGetter.cli()

pyobo/sources/nlm/nlm_publisher.py

@@ -0,0 +1,36 @@
+"""Converter for NLM Providers."""
+
+from collections.abc import Iterable
+
+from pyobo.sources.nlm.utils import PREFIX_PUBLISHER, PUBLISHER_TERM, get_publishers
+from pyobo.struct import CHARLIE_TERM, HUMAN_TERM, Obo, Term
+
+__all__ = [
+    "NLMPublisherGetter",
+]
+
+
+class NLMPublisherGetter(Obo):
+    """An ontology representation of NLM Publishers."""
+
+    bioversions_key = ontology = PREFIX_PUBLISHER
+    dynamic_version = True
+    root_terms = [PUBLISHER_TERM.reference]
+
+    def iter_terms(self, force: bool = False) -> Iterable[Term]:
+        """Iterate over gene terms for NLM Catalog."""
+        yield from get_terms(force=force)
+
+
+def get_terms(*, force: bool = False) -> Iterable[Term]:
+    """Get NLM publisher terms."""
+    yield PUBLISHER_TERM
+    yield CHARLIE_TERM
+    yield HUMAN_TERM
+
+    journal_id_to_publisher_key = get_publishers(force=force)
+    yield from sorted(set(journal_id_to_publisher_key.values()))
+
+
+if __name__ == "__main__":
+    NLMPublisherGetter.cli()

pyobo/sources/nlm/utils.py

@@ -0,0 +1,116 @@
+"""Utilities for NLM."""
+
+from collections.abc import Iterable
+from xml.etree import ElementTree
+
+from tqdm import tqdm
+
+from pyobo import Reference, Term, TypeDef, default_reference, ensure_path
+from pyobo.struct.struct import CHARLIE_TERM, PYOBO_INJECTED
+from pyobo.struct.typedef import has_end_date, has_start_date
+from pyobo.utils.path import ensure_df
+
+PREFIX_CATALOG = "nlm"
+PREFIX_PUBLISHER = "nlm.publisher"
+
+CATALOG_TO_PUBLISHER = "https://ftp.ncbi.nlm.nih.gov/pubmed/xmlprovidernames.txt"
+JOURNAL_INFO_PATH = "https://ftp.ncbi.nlm.nih.gov/pubmed/jourcache.xml"
+PUBLISHED_IN = TypeDef(
+    reference=default_reference(PREFIX_CATALOG, "published_in", name="published in"),
+    xrefs=[
+        Reference(prefix="biolink", identifier="published_in"),
+        Reference(prefix="uniprot.core", identifier="publishedIn"),
+    ],
+)
+JOURNAL_TERM = (
+    Term(reference=default_reference(PREFIX_CATALOG, "journal", name="journal"))
+    .append_exact_match(Reference(prefix="SIO", identifier="000160"))
+    .append_exact_match(Reference(prefix="FBCV", identifier="0000787"))
+    .append_exact_match(Reference(prefix="MI", identifier="0885"))
+    .append_exact_match(Reference(prefix="bibo", identifier="Journal"))
+    .append_exact_match(Reference(prefix="uniprot.core", identifier="Journal"))
+    .append_contributor(CHARLIE_TERM)
+    .append_comment(PYOBO_INJECTED)
+)
+PUBLISHER_TERM = (
+    Term(reference=default_reference(PREFIX_CATALOG, "publisher", name="publisher"))
+    .append_exact_match(Reference(prefix="biolink", identifier="publisher"))
+    .append_exact_match(Reference(prefix="schema", identifier="publisher"))
+    .append_exact_match(Reference(prefix="uniprot.core", identifier="publisher"))
+    .append_contributor(CHARLIE_TERM)
+    .append_comment(PYOBO_INJECTED)
+)
+
+
+def get_publishers(*, force: bool = False) -> dict[str, Term]:
+    """Get NLM publishers."""
+    journal_to_publisher_df = ensure_df(
+        PREFIX_CATALOG, url=CATALOG_TO_PUBLISHER, sep="|", force=force, dtype=str
+    )
+    journal_id_to_publisher_key: dict[str, Term] = {
+        journal_id: Term(
+            reference=Reference(prefix=PREFIX_PUBLISHER, identifier=identifier, name=name),
+            type="Instance",
+        ).append_parent(PUBLISHER_TERM)
+        for journal_id, identifier, name in journal_to_publisher_df.values
+    }
+    return journal_id_to_publisher_key
+
+
+def get_journals(
+    *, force: bool = False, journal_id_to_publisher_key: dict[str, Term] | None = None
+) -> Iterable[Term]:
+    """Get NLM Catalog terms."""
+    path = ensure_path(PREFIX_CATALOG, url=JOURNAL_INFO_PATH, force=force)
+    root = ElementTree.parse(path).getroot()
+
+    if journal_id_to_publisher_key is None:
+        journal_id_to_publisher_key = get_publishers(force=force)
+    elements = root.findall("Journal")
+    for element in elements:
+        if term := _process_journal(element, journal_id_to_publisher_key):
+            yield term
+
+
+def _process_journal(element, journal_id_to_publisher_key: dict[str, Term]) -> Term | None:
+    # TODO enrich with context from https://ftp.ncbi.nlm.nih.gov/pubmed/J_Entrez.txt and https://ftp.ncbi.nlm.nih.gov/pubmed/J_Medline.txt
+
+    nlm_id = element.findtext("NlmUniqueID")
+    name = element.findtext("Name")
+
+    if not nlm_id.isnumeric():
+        # TODO investigate these records, which all appear to have IDs that
+        #  end in R like 17410670R (Proceedings of the staff meetings. Honolulu. Clinic)
+        #  which corresponds to https://www.ncbi.nlm.nih.gov/nlmcatalog/287649
+        return None
+
+    issns = [(issn.text, issn.attrib["type"]) for issn in element.findall("Issn")]
+    # ActivityFlag is either "0" or "1"
+    term = Term(
+        reference=Reference(prefix=PREFIX_CATALOG, identifier=nlm_id, name=name),
+        type="Instance",
+    )
+    term.append_parent(JOURNAL_TERM)
+    for synonym in element.findall("Alias"):
+        term.append_synonym(synonym.text)
+    for issn, _issn_type in issns:
+        if issn.isnumeric():
+            issn = issn[:4] + "-" + issn[4:]
+
+        # TODO include ISSN type, this is important
+        #  to determine a "canonical" one
+        term.append_xref(Reference(prefix="issn", identifier=issn))
+    if start_year := element.findtext("StartYear"):
+        if len(start_year) != 4:
+            tqdm.write(f"[{term.curie}] invalid start year: {start_year}")
+        else:
+            term.annotate_year(has_start_date, start_year)
+    if end_year := element.findtext("EndYear"):
+        if len(end_year) != 4:
+            tqdm.write(f"[{term.curie}] invalid end year: {end_year}")
+        else:
+            term.annotate_year(has_end_date, end_year)
+    # FIXME this whole thing needs reinvestigating
+    if publisher_reference := journal_id_to_publisher_key.get(term.identifier):
+        term.annotate_object(PUBLISHED_IN, publisher_reference.reference)
+    return term

pyobo/sources/npass.py

@@ -6,7 +6,7 @@ from collections.abc import Iterable
 import pandas as pd
 from tqdm.auto import tqdm
 
-from ..struct import Obo, Reference, Synonym, Term
+from ..struct import Obo, Reference, Term
 from ..utils.path import ensure_df
 
 __all__ = [
@@ -32,11 +32,6 @@ class NPASSGetter(Obo):
         return iter_terms(force=force, version=self._version_or_raise)
 
 
-def get_obo(force: bool = False) -> Obo:
-    """Get NPASS as OBO."""
-    return NPASSGetter()
-
-
 def get_df(version: str, force: bool = False) -> pd.DataFrame:
     """Get the NPASS chemical nomenclature."""
     base_url = f"https://bidd.group/NPASS/downloadFiles/NPASSv{version}_download"
@@ -71,7 +66,10 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
         # TODO check that the first is always the parent compound?
         if pd.notna(pubchem_compound_ids):
             pubchem_compound_ids = [
-                yy.strip() for xx in pubchem_compound_ids.split(";") for yy in xx.strip().split(",")
+                zz
+                for xx in pubchem_compound_ids.split(";")
+                for yy in xx.strip().split(",")
+                if (zz := yy.strip())
             ]
             if len(pubchem_compound_ids) > 1:
                 logger.debug("multiple cids for %s: %s", identifier, pubchem_compound_ids)
@@ -82,7 +80,7 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
 
         for synonym in [iupac]:
             if pd.notna(synonym):
-                term.append_synonym(Synonym(name=synonym))
+                term.append_synonym(synonym)
 
         yield term
 

pyobo/sources/omim_ps.py

@@ -11,7 +11,6 @@ __all__ = [
     "OMIMPSGetter",
 ]
 
-
 logger = logging.getLogger(__name__)
 PREFIX = "omim.ps"
 URL = "https://omim.org/phenotypicSeriesTitles/all"
@@ -25,8 +24,16 @@ class OMIMPSGetter(Obo):
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""
         soup = get_soup(URL, user_agent="Mozilla/5.0")
-        rows = soup.find(id="mimContent").find("table").find("tbody").find_all("tr")
-        for row in rows:
+        content = soup.find(id="mimContent")
+        if content is None:
+            raise ValueError
+        table = content.find("table")  # type:ignore[attr-defined]
+        if table is None:
+            raise ValueError
+        tbody = table.find("tbody")
+        if tbody is None:
+            raise ValueError
+        for row in tbody.find_all("tr"):
             anchor = row.find("td").find("a")
             name = anchor.text.strip()
             identifier = anchor.attrs["href"][len("/phenotypicSeries/") :]

pyobo/sources/pathbank.py

@@ -5,6 +5,7 @@ from __future__ import annotations
 import logging
 from collections import defaultdict
 from collections.abc import Iterable, Mapping
+from itertools import chain
 
 import pandas as pd
 from tqdm.auto import tqdm
@@ -77,11 +78,6 @@ class PathBankGetter(Obo):
         return iter_terms(force=force, version=self._version_or_raise)
 
 
-def get_obo(force: bool = False) -> Obo:
-    """Get PathBank as OBO."""
-    return PathBankGetter(force=force)
-
-
 def get_proteins_df(version: str, force: bool = False) -> pd.DataFrame:
     """Get the proteins dataframe."""
     proteins_df = ensure_df(
@@ -165,9 +161,9 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
             #  but there are weird parser errors
         )
         term.append_exact_match(Reference(prefix="smpdb", identifier=smpdb_id))
-        term.append_property(has_category, subject.lower().replace(" ", "_"))
-        term.extend_relationship(has_participant, smpdb_id_to_proteins[smpdb_id])
-        term.extend_relationship(has_participant, smpdb_id_to_metabolites[smpdb_id])
+        term.annotate_string(has_category, subject.lower().replace(" ", "_"))
+        for participant in chain(smpdb_id_to_proteins[smpdb_id], smpdb_id_to_metabolites[smpdb_id]):
+            term.append_relationship(has_participant, participant)
         yield term
 
 

pyobo/sources/pfam/__init__.py

@@ -0,0 +1,9 @@
+"""Resources from PFAM."""
+
+from .pfam import PfamGetter
+from .pfam_clan import PfamClanGetter
+
+__all__ = [
+    "PfamClanGetter",
+    "PfamGetter",
+]

pyobo/sources/{pfam.py → pfam/pfam.py}

@@ -4,8 +4,8 @@ from collections.abc import Iterable
 
 import pandas as pd
 
-from ..struct import Obo, Reference, Term
-from ..utils.path import ensure_df
+from ...struct import Obo, Reference, Term
+from ...utils.path import ensure_df
 
 __all__ = [
     "PfamGetter",
@@ -47,11 +47,6 @@ class PfamGetter(Obo):
         return iter_terms(self._version_or_raise, force=force)
 
 
-def get_obo(force: bool = False) -> Obo:
-    """Get PFAM as OBO."""
-    return PfamGetter(force=force)
-
-
 def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
     """Iterate PFAM terms."""
     df = get_pfam_clan_df(version=version, force=force)
@@ -67,4 +62,4 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
 
 
 if __name__ == "__main__":
-    get_obo().write_default()
+    PfamGetter.cli()

pyobo/sources/{pfam_clan.py → pfam/pfam_clan.py}

@@ -5,7 +5,7 @@ from collections.abc import Iterable
 from tqdm.auto import tqdm
 
 from .pfam import get_pfam_clan_df
-from ..struct import Obo, Reference, Term
+from ...struct import Obo, Reference, Term
 
 __all__ = [
     "PfamClanGetter",
@@ -25,11 +25,6 @@ class PfamClanGetter(Obo):
         return iter_terms(version=self._version_or_raise, force=force)
 
 
-def get_obo(force: bool = False) -> Obo:
-    """Get PFAM Clans as OBO."""
-    return PfamClanGetter(force=force)
-
-
 # TODO could get definitions from ftp://ftp.ebi.ac.uk/pub/databases/Pfam/releases/Pfam33.0/Pfam-C.gz
 
 
@@ -46,4 +41,4 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
 
 
 if __name__ == "__main__":
-    get_obo().write_default()
+    PfamClanGetter.cli()