pyobo 0.11.0__py3-none-any.whl → 0.11.2__py3-none-any.whl

pyobo/sources/complexportal.py

@@ -58,11 +58,31 @@ def _parse_members(s) -> list[tuple[Reference, str]]:
     for member in s.split("|"):
         entity_id, count = member.split("(")
         count = count.rstrip(")")
-        if ":" in entity_id:
+        if entity_id.startswith("URS"):
+            prefix, identifier = "rnacentral", entity_id
+        elif entity_id.startswith("CPX"):
+            # TODO why self xref?
+            prefix, identifier = "complexportal", entity_id
+        elif entity_id.startswith("["):
+            continue  # this is a list of uniprot IDs, not sure what to do with this
+        elif entity_id.startswith("EBI-"):
+            continue
+        elif ":" not in entity_id:
+            if "PRO_" in entity_id:
+                prefix = "uniprot.chain"
+                identifier = entity_id.split("-")[1]
+            elif "-" in entity_id:
+                prefix, identifier = "uniprot.isoform", entity_id
+            else:
+                prefix, identifier = "uniprot", entity_id
+        else:
             prefix, identifier = entity_id.split(":", 1)
+        try:
+            reference = Reference(prefix=prefix, identifier=identifier)
+        except ValueError:
+            tqdm.write(f"failed to validate reference: {entity_id}")
         else:
-            prefix, identifier = "uniprot", entity_id
-        rv.append((Reference(prefix=prefix, identifier=identifier), count))
+            rv.append((reference, count))
     return rv
 
 
@@ -74,27 +94,40 @@ def _parse_xrefs(s) -> list[tuple[Reference, str]]:
     for xref in s.split("|"):
         xref = xref.replace("protein ontology:PR:", "PR:")
         xref = xref.replace("protein ontology:PR_", "PR:")
+        xref = xref.replace("rhea:rhea ", "rhea:")
+        xref = xref.replace("rhea:Rhea ", "rhea:")
+        xref = xref.replace("rhea:RHEA:rhea", "rhea:")
+        xref = xref.replace("rhea:RHEA: ", "rhea:")
+        xref = xref.replace("rhea:RHEA:rhea ", "rhea:")
+        xref = xref.replace("intenz:RHEA:", "rhea:")
+        xref = xref.replace("eccode::", "eccode:")
+        xref = xref.replace("eccode:EC:", "eccode:")
+        xref = xref.replace("intenz:EC:", "eccode:")
+        xref = xref.replace("eccode:RHEA:", "rhea:")
+        xref = xref.replace("efo:MONDO:", "MONDO:")
+        xref = xref.replace("omim:MIM:", "omim:")
+        xref = xref.replace("efo:HP:", "HP:")
+        xref = xref.replace("efo:Orphanet:", "Orphanet:")
+        xref = xref.replace("orphanet:ORDO:", "Orphanet:")
+        xref = xref.replace("biorxiv:doi.org/", "doi:")
+        xref = xref.replace("emdb:EMDB-", "emdb:EMD-")
+        xref = xref.replace("wwpdb:EMD-", "emdb:EMD-")
+        xref = xref.replace("signor:CPX-", "complexportal:CPX-")
+
         try:
             xref_curie, note = xref.split("(")
         except ValueError:
             logger.warning("xref missing (: %s", xref)
             continue
         note = note.rstrip(")")
-        note.replace("rhea:rhea ", "rhea:")
-        note.replace("rhea:Rhea ", "rhea:")
-        note.replace("eccode::", "eccode:")
-        note.replace("eccode:EC:", "eccode:")
-        note.replace("eccode:RHEA:", "rhea:")
-        if note.lower().startswith("rhea "):
-            note = note[len("Rhea ") :]
-            if note.lower().startswith("rhea:rhea "):
-                note = note[len("rhea:rhea ") :]
-        if note.lower().startswith("EC:"):
-            note = note[len("EC:") :]
+
+        if xref_curie.startswith("intenz:"):
+            xref_curie = _clean_intenz(xref_curie)
+
         try:
             reference = Reference.from_curie(xref_curie)
         except ValueError:
-            logger.warning("can not parse CURIE: %s", xref)
+            logger.warning("can not parse CURIE: %s", xref_curie)
             continue
         if reference is None:
             logger.warning("reference is None after parsing: %s", xref)
@@ -103,6 +136,12 @@ def _parse_xrefs(s) -> list[tuple[Reference, str]]:
     return rv
 
 
+def _clean_intenz(s: str) -> str:
+    for _ in range(3):
+        s = s.rstrip("-").rstrip(".")
+    return s
+
+
 class ComplexPortalGetter(Obo):
     """An ontology representation of the Complex Portal."""
 

pyobo/sources/dictybase_gene.py

@@ -9,8 +9,7 @@ from collections.abc import Iterable
 import pandas as pd
 from tqdm.auto import tqdm
 
-from pyobo.struct import Obo, Reference, Synonym, Term, from_species, has_gene_product
-from pyobo.utils.io import multisetdict
+from pyobo.struct import Obo, Synonym, Term, from_species, has_gene_product
 from pyobo.utils.path import ensure_df
 
 __all__ = [
@@ -49,10 +48,11 @@ def get_obo(force: bool = False) -> Obo:
 
 def get_terms(force: bool = False) -> Iterable[Term]:
     """Get terms."""
+    # TODO the mappings file has actually no uniprot at all, and requires text mining
     # DDB ID	DDB_G ID	Name	UniProt ID
-    uniprot_mappings = multisetdict(
-        ensure_df(PREFIX, url=URL, force=force, name="uniprot_mappings.tsv", usecols=[1, 3]).values
-    )
+    # uniprot_mappings = multisetdict(
+    #     ensure_df(PREFIX, url=URL, force=force, name="uniprot_mappings.tsv", usecols=[1, 3]).values
+    # )
 
     terms = ensure_df(PREFIX, url=URL, force=force, name="gene_info.tsv")
     # GENE ID (DDB_G ID)	Gene Name	Synonyms	Gene products
@@ -68,10 +68,15 @@ def get_terms(force: bool = False) -> Iterable[Term]:
         if synonyms and pd.notna(synonyms):
             for synonym in synonyms.split(","):
                 term.append_synonym(Synonym(synonym.strip()))
-        for uniprot_id in uniprot_mappings.get(identifier, []):
-            if not uniprot_id or pd.isna(uniprot_id) or uniprot_id not in {"unknown", "pseudogene"}:
-                continue
-            term.append_relationship(has_gene_product, Reference.auto("uniprot", uniprot_id))
+        # for uniprot_id in uniprot_mappings.get(identifier, []):
+        #     if not uniprot_id or pd.isna(uniprot_id) or uniprot_id in {"unknown", "pseudogene"}:
+        #         continue
+        #     try:
+        #         uniprot_ref = Reference(prefix="uniprot", identifier=uniprot_id)
+        #     except ValueError:
+        #         tqdm.write(f"[dictybase.gene] invalid uniprot ref: {uniprot_id}")
+        #     else:
+        #         term.append_relationship(has_gene_product, uniprot_ref)
 
         term.set_species(identifier="44689", name="Dictyostelium discoideum")
         yield term

pyobo/sources/drugcentral.py

@@ -68,6 +68,9 @@ def iter_terms() -> Iterable[Term]:
                 if xref_prefix_norm is None:
                     tqdm.write(f"did not normalize {prefix}:{identifier}")
                     continue
+                if xref_prefix_norm == "pdb.ligand":
+                    # there is a weird invalid escaped \W appearing in pdb ligand ids
+                    identifier = identifier.strip()
                 identifier = bioregistry.standardize_identifier(xref_prefix_norm, identifier)
                 xrefs[str(drugcentral_id)].append(
                     Reference(prefix=xref_prefix_norm, identifier=identifier)
@@ -98,4 +101,4 @@ def iter_terms() -> Iterable[Term]:
 
 
 if __name__ == "__main__":
-    get_obo().write_default(write_obo=True)
+    DrugCentralGetter.cli()

pyobo/sources/expasy.py

@@ -1,6 +1,7 @@
 """Convert ExPASy to OBO."""
 
 import logging
+import re
 from collections import defaultdict
 from collections.abc import Iterable, Mapping
 from typing import Any, Optional
@@ -42,7 +43,7 @@ class ExpasyGetter(Obo):
     """A getter for ExPASy Enzyme Classes."""
 
     bioversions_key = ontology = PREFIX
-    typedefs = [has_member, enables]
+    typedefs = [has_member, enables, term_replaced_by]
     root_terms = [
         Reference(prefix="eccode", identifier="1"),
         Reference(prefix="eccode", identifier="2"),
@@ -145,7 +146,9 @@ def get_terms(version: str, force: bool = False) -> Iterable[Term]:
         for domain in data.get("domains", []):
             term.append_relationship(
                 has_member,
-                Reference(prefix=domain["namespace"], identifier=domain["identifier"]),
+                Reference.model_validate(
+                    {"prefix": domain["namespace"], "identifier": domain["identifier"]},
+                ),
             )
         for protein in data.get("proteins", []):
             term.append_relationship(
@@ -248,8 +251,10 @@ def get_database(lines: Iterable[str]) -> Mapping:
             elif descriptor == DE and value == "Deleted entry.":
                 ec_data_entry["deleted"] = True
             elif descriptor == DE and value.startswith("Transferred entry: "):
-                value = value[len("Transferred entry: ") :].rstrip().rstrip(".")
-                ec_data_entry["transfer_id"] = value.split(" and ")
+                # TODO There's a situation where there are enough transfers that it goes on to a second line
+                #  the following line just gives up on this one. or maybe I don't understand
+                value = value.strip().removesuffix("and").rstrip(",").strip()
+                ec_data_entry["transfer_id"] = _parse_transfer(value)
             elif descriptor == DE:
                 ec_data_entry["concept"]["name"] = value.rstrip(".")  # type:ignore
             elif descriptor == AN:
@@ -279,6 +284,19 @@ def get_database(lines: Iterable[str]) -> Mapping:
     return rv
 
 
+TRANSFER_SPLIT_RE = re.compile(r",\s*|\s+and\s+")
+
+
+def _parse_transfer(value: str) -> list[str]:
+    """Parse transferred entry string.
+
+    >>> _parse_transfer("Transferred entry: 1.1.1.198, 1.1.1.227 and 1.1.1.228.")
+    ['1.1.1.198', '1.1.1.227', '1.1.1.228']
+    """
+    value = value[len("Transferred entry: ") :].rstrip().rstrip(".")
+    return sorted(x.strip().removeprefix("and").strip() for x in TRANSFER_SPLIT_RE.split(value))
+
+
 def _group_by_id(lines):
     """Group lines by identifier."""
     groups = []

pyobo/sources/flybase.py

@@ -7,6 +7,7 @@ import pandas as pd
 from tqdm.auto import tqdm
 
 from pyobo import Reference
+from pyobo.resources.so import get_so_name
 from pyobo.struct import Obo, Term, from_species, orthologous
 from pyobo.utils.io import multisetdict
 from pyobo.utils.path import ensure_df
@@ -133,7 +134,7 @@ def get_terms(version: str, force: bool = False) -> Iterable[Term]:
                 "FlyBase gene type is missing mapping to Sequence Ontology (SO): %s", gtype
             )
         else:
-            so[gtype] = Reference.auto("SO", so_id)
+            so[gtype] = Reference(prefix="SO", identifier=so_id, name=get_so_name(so_id))
 
     for _, reference in sorted(so.items()):
         yield Term(reference=reference)
@@ -153,7 +154,7 @@ def get_terms(version: str, force: bool = False) -> Iterable[Term]:
         for hgnc_curie in human_orthologs.get(identifier, []):
             if not hgnc_curie or pd.isna(hgnc_curie):
                 continue
-            hgnc_ortholog = Reference.from_curie(hgnc_curie, auto=True)
+            hgnc_ortholog = Reference.from_curie(hgnc_curie)
             if hgnc_ortholog is None:
                 tqdm.write(f"[{PREFIX}] {identifier} had invalid ortholog: {hgnc_curie}")
             else:

pyobo/sources/hgnc.py

@@ -13,6 +13,7 @@ from tabulate import tabulate
 from tqdm.auto import tqdm
 
 from pyobo.api.utils import get_version
+from pyobo.resources.so import get_so_name
 from pyobo.struct import (
     Obo,
     Reference,
@@ -37,8 +38,8 @@ logger = logging.getLogger(__name__)
 
 PREFIX = "hgnc"
 DEFINITIONS_URL_FMT = (
-    "http://ftp.ebi.ac.uk/pub/databases/genenames/hgnc/"
-    "archive/monthly/json/hgnc_complete_set_{version}.json"
+    "https://storage.googleapis.com/public-download-files/hgnc/archive/archive/monthly/json/"
+    "hgnc_complete_set_{version}.json"
 )
 
 previous_symbol_type = SynonymTypeDef.from_text("previous_symbol")
@@ -222,7 +223,7 @@ class HGNCGetter(Obo):
         alias_symbol_type,
     ]
     root_terms = [
-        Reference(prefix="so", identifier=so_id)
+        Reference(prefix="SO", identifier=so_id, name=get_so_name(so_id))
         for so_id in sorted(set(LOCUS_TYPE_TO_SO.values()))
         if so_id
     ]
@@ -256,7 +257,7 @@ def get_terms(version: Optional[str] = None, force: bool = False) -> Iterable[Te
     yield Term.from_triple("NCBITaxon", "9606", "Homo sapiens")
     yield from sorted(
         {
-            Term(reference=Reference.auto("SO", so_id))
+            Term(reference=Reference(prefix="SO", identifier=so_id, name=get_so_name(so_id)))
             for so_id in sorted(LOCUS_TYPE_TO_SO.values())
             if so_id
         },
@@ -363,23 +364,25 @@ def get_terms(version: Optional[str] = None, force: bool = False) -> Iterable[Te
             xref_identifiers = entry.pop(key, None)
             if xref_identifiers is None:
                 continue
-
             if isinstance(xref_identifiers, (str, int)):
+                xref_identifiers = [str(xref_identifiers)]
+
+            if xref_prefix == "merops.entry":
+                continue
+                # e.g., XM02-001 should be rewritten as XM02.001
+                xref_identifiers = [i.replace("-", ".") for i in xref_identifiers]
+
+            if xref_prefix == "refseq":
+                # e.g., strip off dots without substantiated record versions like in NM_021728.
+                xref_identifiers = [i.strip(".") for i in xref_identifiers]
+
+            if len(xref_identifiers) == 1:
                 term.append_exact_match(
-                    Reference(prefix=xref_prefix, identifier=str(xref_identifiers))
+                    Reference(prefix=xref_prefix, identifier=str(xref_identifiers[0]))
                 )
-            elif isinstance(xref_identifiers, list):
-                if len(xref_identifiers) == 1:
-                    term.append_exact_match(
-                        Reference(prefix=xref_prefix, identifier=str(xref_identifiers[0]))
-                    )
-                else:
-                    for xref_identifier in xref_identifiers:
-                        term.append_xref(
-                            Reference(prefix=xref_prefix, identifier=str(xref_identifier))
-                        )
             else:
-                raise TypeError
+                for xref_identifier in xref_identifiers:
+                    term.append_xref(Reference(prefix=xref_prefix, identifier=str(xref_identifier)))
 
         for pubmed_id in entry.pop("pubmed_id", []):
             term.append_provenance(Reference(prefix="pubmed", identifier=str(pubmed_id)))
@@ -416,9 +419,11 @@ def get_terms(version: Optional[str] = None, force: bool = False) -> Iterable[Te
         locus_group = entry.pop("locus_group")
         so_id = LOCUS_TYPE_TO_SO.get(locus_type)
         if so_id:
-            term.append_parent(Reference.auto("SO", so_id))
+            term.append_parent(Reference(prefix="SO", identifier=so_id, name=get_so_name(so_id)))
         else:
-            term.append_parent(Reference.auto("SO", "0000704"))  # gene
+            term.append_parent(
+                Reference(prefix="SO", identifier="0000704", name=get_so_name("0000704"))
+            )  # gene
             unhandle_locus_types[locus_type][identifier] = term
             term.append_property("locus_type", locus_type)
             term.append_property("locus_group", locus_group)

pyobo/sources/hgncgenefamily.py

@@ -21,13 +21,13 @@ __all__ = [
 ]
 
 PREFIX = "hgnc.genegroup"
-FAMILIES_URL = "ftp://ftp.ebi.ac.uk/pub/databases/genenames/new/csv/genefamily_db_tables/family.csv"
+FAMILIES_URL = "https://storage.googleapis.com/public-download-files/hgnc/csv/csv/genefamily_db_tables/family.csv"
 # TODO use family_alias.csv
-HIERARCHY_URL = (
-    "ftp://ftp.ebi.ac.uk/pub/databases/genenames/new/csv/genefamily_db_tables/hierarchy.csv"
-)
+HIERARCHY_URL = "https://storage.googleapis.com/public-download-files/hgnc/csv/csv/genefamily_db_tables/hierarchy.csv"
 
-symbol_type = SynonymTypeDef.from_text("symbol")
+symbol_type = SynonymTypeDef(
+    reference=Reference(prefix="OMO", identifier="0004000", name="has symbol")
+)
 
 
 class HGNCGroupGetter(Obo):
@@ -78,7 +78,7 @@ def get_terms(force: bool = False) -> Iterable[Term]:
                     name=parent.name,
                 )
             )
-    gene_group = Reference.auto("SO", "0005855")
+    gene_group = Reference(prefix="SO", identifier="0005855", name="gene group")
     yield Term(reference=gene_group)
     for term in terms:
         if not term.parents:
@@ -98,7 +98,7 @@ def _get_terms_helper(force: bool = False) -> Iterable[Term]:
             definition=definition,
         )
         if pubmed_ids and pd.notna(pubmed_ids):
-            for s in pubmed_ids.split(","):
+            for s in pubmed_ids.replace(" ", ",").split(","):
                 term.append_provenance(Reference(prefix="pubmed", identifier=s.strip()))
         if desc_go and pd.notna(desc_go):
             go_id = desc_go[len("http://purl.uniprot.org/go/") :]

pyobo/sources/kegg/genome.py

@@ -3,6 +3,8 @@
 Run with ``python -m pyobo.sources.kegg.genome``
 """
 
+from __future__ import annotations
+
 import logging
 from collections.abc import Iterable
 
@@ -46,8 +48,11 @@ def get_obo() -> Obo:
     return KEGGGenomeGetter()
 
 
-def parse_genome_line(line: str) -> KEGGGenome:
+def parse_genome_line(line: str) -> KEGGGenome | None:
     """Parse a line from the KEGG Genome database."""
+    if not line.startswith("T"):
+        #  This is for an NCBI Taxonomy
+        return None
     line = line.strip()
     identifier, rest = _s(line, "\t")
     identifier = identifier[len("gn:") :]
@@ -94,6 +99,8 @@ def iter_kegg_genomes(version: str, desc: str) -> Iterable[KEGGGenome]:
     it = tqdm(lines, desc=desc, unit_scale=True, unit="genome")
     for line in it:
         yv = parse_genome_line(line)
+        if yv is None:
+            continue
         it.set_postfix({"id": yv.identifier, "name": yv.name})
         yield yv
 
@@ -105,11 +112,16 @@ def iter_terms(version: str) -> Iterable[Term]:
     for kegg_genome in iter_kegg_genomes(version=version, desc="KEGG Genomes"):
         if kegg_genome.identifier in SKIP:
             continue
-        term = Term.from_triple(
-            prefix=KEGG_GENOME_PREFIX,
-            identifier=kegg_genome.identifier,
-            name=kegg_genome.name,
-        )
+
+        try:
+            reference = Reference(
+                prefix=KEGG_GENOME_PREFIX, identifier=kegg_genome.identifier, name=kegg_genome.name
+            )
+        except ValueError:
+            tqdm.write(f"[{KEGG_GENOME_PREFIX}] invalid identifier: {kegg_genome}")
+            continue
+
+        term = Term(reference=reference)
         if kegg_genome.taxonomy_id is not None:
             taxonomy_name = get_ncbitaxon_name(kegg_genome.taxonomy_id)
             if taxonomy_name is None:

pyobo/sources/mirbase.py

@@ -136,9 +136,15 @@ def _process_definitions_lines(
             xref_prefix = xref_mapping.get(xref_prefix, xref_prefix)
             if xref_prefix == "pictar":
                 continue
-            xrefs.append(
-                Reference(prefix=xref_prefix, identifier=xref_identifier, name=xref_label or None)
-            )
+
+            try:
+                xref = Reference(
+                    prefix=xref_prefix, identifier=xref_identifier, name=xref_label or None
+                )
+            except ValueError:
+                tqdm.write(f"invalid xref: {xref_prefix}:{xref_identifier}")
+            else:
+                xrefs.append(xref)
 
         # TODO add pubmed references
 

pyobo/sources/npass.py

@@ -39,7 +39,7 @@ def get_obo(force: bool = False) -> Obo:
 
 def get_df(version: str, force: bool = False) -> pd.DataFrame:
     """Get the NPASS chemical nomenclature."""
-    base_url = f"http://bidd.group/NPASS/downloadFiles/NPASSv{version}_download"
+    base_url = f"https://bidd.group/NPASS/downloadFiles/NPASSv{version}_download"
     url = f"{base_url}_naturalProducts_generalInfo.txt"
     return ensure_df(
         PREFIX,

pyobo/sources/pathbank.py

@@ -1,5 +1,7 @@
 """Converter for PathBank."""
 
+from __future__ import annotations
+
 import logging
 from collections import defaultdict
 from collections.abc import Iterable, Mapping
@@ -8,7 +10,7 @@ import pandas as pd
 from tqdm.auto import tqdm
 
 from ..struct import Obo, Reference, Term
-from ..struct.typedef import has_participant
+from ..struct.typedef import has_category, has_participant
 from ..utils.path import ensure_df
 
 __all__ = [
@@ -68,7 +70,7 @@ class PathBankGetter(Obo):
     """An ontology representation of PathBank's pathway nomenclature."""
 
     ontology = bioversions_key = PREFIX
-    typedefs = [has_participant]
+    typedefs = [has_participant, has_category]
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:
         """Iterate over terms in the ontology."""
@@ -103,21 +105,30 @@ def get_protein_mapping(version: str, force: bool = False) -> Mapping[str, set[R
     for pathway_id, protein_id in tqdm(
         proteins_df.values, desc=f"[{PREFIX}] mapping proteins", unit_scale=True
     ):
-        # TODO get protein names
-        smpdb_id_to_proteins[pathway_id].add(Reference(prefix="uniprot", identifier=protein_id))
+        try:
+            if "-" in protein_id:
+                reference = Reference(prefix="uniprot.isoform", identifier=protein_id)
+            else:
+                reference = Reference(prefix="uniprot", identifier=protein_id)
+        except ValueError:
+            tqdm.write(f"[pathbank] invalid uniprot identifier: {protein_id}")
+        else:
+            smpdb_id_to_proteins[pathway_id].add(reference)
     return smpdb_id_to_proteins
 
 
 def get_metabolite_df(version: str, force: bool = False) -> pd.DataFrame:
     """Get the metabolites dataframe."""
-    return ensure_df(
+    df = ensure_df(
         PREFIX,
         url=METABOLITE_URL,
         sep=",",
-        usecols=["PathBank ID", "Metabolite ID", "Metabolite Name"],
+        usecols=["PathBank ID", "ChEBI ID"],
         force=force,
         version=version,
     )
+    df = df[df["ChEBI ID"].notna()]
+    return df
 
 
 def get_metabolite_mapping(version: str, force: bool = False) -> Mapping[str, set[Reference]]:
@@ -125,17 +136,20 @@ def get_metabolite_mapping(version: str, force: bool = False) -> Mapping[str, se
     metabolites_df = get_metabolite_df(version=version, force=force)
     smpdb_id_to_metabolites = defaultdict(set)
     it = tqdm(metabolites_df.values, desc=f"[{PREFIX}] mapping metabolites", unit_scale=True)
-    for pathway_id, metabolite_id, metabolite_name in it:
-        smpdb_id_to_metabolites[pathway_id].add(
-            Reference(
-                prefix=PREFIX,
-                identifier=metabolite_id,
-                name=metabolite_name,
-            )
-        )
+    for pathway_id, metabolite_id in it:
+        reference = Reference(prefix="chebi", identifier=metabolite_id.strip())
+        smpdb_id_to_metabolites[pathway_id].add(reference)
     return smpdb_id_to_metabolites
 
 
+def _clean_description(description: str) -> str | None:
+    """Clean the description."""
+    if pd.isna(description) or not description:
+        return None
+    parts = [part.strip() for part in description.strip().splitlines()]
+    return " ".join(parts)
+
+
 def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
     """Get PathBank's terms."""
     smpdb_id_to_proteins = get_protein_mapping(version=version, force=force)
@@ -147,16 +161,11 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
         reference = Reference(prefix=PREFIX, identifier=pathbank_id, name=name)
         term = Term(
             reference=reference,
-            # definition=description.replace('\n', ' '),
-            xrefs=[Reference(prefix="smpdb", identifier=smpdb_id)],
-        )
-        term.append_parent(
-            Reference(
-                prefix=PREFIX,
-                identifier=subject.lower().replace(" ", "_"),
-                name=subject,
-            )
+            # TODO use _clean_description(description) to add a description,
+            #  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])
         yield term

pyobo/sources/pombase.py

@@ -9,6 +9,7 @@ from tqdm.auto import tqdm
 
 import pyobo
 from pyobo import Reference
+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
 
@@ -19,7 +20,7 @@ __all__ = [
 logger = logging.getLogger(__name__)
 
 PREFIX = "pombase"
-URL = "https://www.pombase.org/data/names_and_identifiers/gene_IDs_names_products.tsv"
+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"
 
 
@@ -68,9 +69,11 @@ def get_terms(version: str, force: bool = False) -> Iterable[Term]:
             if hgnc_id is not None:
                 identifier_to_hgnc_ids[identifier].add(hgnc_id)
 
-    df = ensure_df(PREFIX, url=URL, force=force, header=None, version=version)
+    df = ensure_df(PREFIX, url=GENE_NAMES_URL, force=force, version=version)
     so = {
-        gtype: Reference.auto("SO", POMBASE_TO_SO[gtype])
+        gtype: Reference(
+            prefix="SO", identifier=POMBASE_TO_SO[gtype], name=get_so_name(POMBASE_TO_SO[gtype])
+        )
         for gtype in sorted(df[df.columns[6]].unique())
     }
     for _, reference in sorted(so.items()):

pyobo/sources/reactome.py

@@ -70,7 +70,9 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
     df["taxonomy_id"] = df["species"].map(get_ncbitaxon_id)
 
     terms = {}
-    it = tqdm(df.values, total=len(df.index), desc=f"mapping {PREFIX}")
+    it = tqdm(
+        df.values, total=len(df.index), desc=f"mapping {PREFIX}", unit_scale=True, unit="pathway"
+    )
     for reactome_id, name, species_name, taxonomy_id in it:
         terms[reactome_id] = term = Term(
             reference=Reference(prefix=PREFIX, identifier=reactome_id, name=name),
@@ -92,10 +94,21 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
         terms[child_id].append_parent(terms[parent_id])
 
     uniprot_pathway_df = ensure_participant_df(version=version, force=force)
-    for uniprot_id, reactome_id in tqdm(uniprot_pathway_df.values, total=len(uniprot_pathway_df)):
-        terms[reactome_id].append_relationship(
-            has_participant, Reference(prefix="uniprot", identifier=uniprot_id)
-        )
+    for uniprot_id, reactome_id in tqdm(
+        uniprot_pathway_df.values,
+        total=len(uniprot_pathway_df),
+        unit_scale=True,
+        unit="pathway-protein",
+    ):
+        if reactome_id not in terms:
+            tqdm.write(f"{reactome_id} appears in uniprot participants file but not pathways file")
+            continue
+
+        if "-" in uniprot_id:
+            reference = Reference(prefix="uniprot.isoform", identifier=uniprot_id)
+        else:
+            reference = Reference(prefix="uniprot", identifier=uniprot_id)
+        terms[reactome_id].append_relationship(has_participant, reference)
 
     chebi_pathway_url = f"https://reactome.org/download/{version}/ChEBI2Reactome_All_Levels.txt"
     chebi_pathway_df = ensure_df(
@@ -106,7 +119,15 @@ def iter_terms(version: str, force: bool = False) -> Iterable[Term]:
         version=version,
         force=force,
     )
-    for chebi_id, reactome_id in tqdm(chebi_pathway_df.values, total=len(chebi_pathway_df)):
+    for chebi_id, reactome_id in tqdm(
+        chebi_pathway_df.values,
+        total=len(chebi_pathway_df),
+        unit_scale=True,
+        unit="pathway-chemical",
+    ):
+        if reactome_id not in terms:
+            tqdm.write(f"{reactome_id} appears in chebi participants file but not pathways file")
+            continue
         terms[reactome_id].append_relationship(
             has_participant, Reference(prefix="chebi", identifier=chebi_id)
         )
@@ -133,4 +154,4 @@ def get_protein_to_pathways() -> Mapping[str, set[str]]:
 
 
 if __name__ == "__main__":
-    get_obo().write_default()
+    ReactomeGetter.cli()