pyobo 0.10.12__py3-none-any.whl → 0.11.1__py3-none-any.whl

pyobo/sources/README.md

@@ -0,0 +1,15 @@
+# Sources
+
+1. Create a new module in `pyobo.sources` named with the prefix for the resource you're ontologizing
+2. Make sure your resource has a corresponding prefix in [the Bioregistry](https://github.com/biopragmatics/bioregistry)
+3. Subclass the `pyobo.Obo` class to represent your resource
+4. Add your resource to the list in `pyobo.sources.__init__`
+
+## What is in scope?
+
+1. Biomedical, semantic web, bibliographic, life sciences, and related natural sciences resources are welcome
+2. The source you want to ontologize should be an identifier resource, i.e., it mints its own identifiers. If you want
+   to ontologize some database that reuses some other identifier resource's identifiers, then this isn't the right
+   place.
+3. Resources that are not possible to download automatically are not in scope for PyOBO. Reproducibility and reusability
+   are core values of this software

pyobo/sources/__init__.py

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 """Sources of OBO content."""
 
 from class_resolver import ClassResolver

pyobo/sources/agrovoc.py

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 """Converter for AGROVOC."""
 
 import pystow
@@ -11,6 +9,8 @@ __all__ = [
     "ensure_agrovoc_graph",
 ]
 
+PREFIX = "agrovoc"
+
 
 def ensure_agrovoc_graph(version: str) -> Graph:
     """Download and parse the given version of AGROVOC."""
@@ -20,5 +20,5 @@ def ensure_agrovoc_graph(version: str) -> Graph:
     graph.bind("skosxl", "http://www.w3.org/2008/05/skos-xl#")
     graph.bind("skos", SKOS)
     graph.bind("dcterms", DCTERMS)
-    graph.bind("agrovoc", "http://aims.fao.org/aos/agrontology#")
+    graph.bind(PREFIX, "http://aims.fao.org/aos/agrontology#")
     return graph

pyobo/sources/antibodyregistry.py

@@ -1,9 +1,8 @@
-# -*- coding: utf-8 -*-
-
 """Converter for the Antibody Registry."""
 
 import logging
-from typing import Iterable, Mapping, Optional
+from collections.abc import Iterable, Mapping
+from typing import Optional
 
 import pandas as pd
 from bioregistry.utils import removeprefix

pyobo/sources/biogrid.py

@@ -1,9 +1,8 @@
-# -*- coding: utf-8 -*-
-
 """Extract and convert BioGRID identifiers."""
 
+from collections.abc import Mapping
 from functools import partial
-from typing import Mapping, Optional
+from typing import Optional
 
 import pandas as pd
 
@@ -77,7 +76,8 @@ def get_ncbigene_mapping() -> Mapping[str, str]:
     .. code-block:: python
 
         from pyobo import get_filtered_xrefs
-        biogrid_ncbigene_mapping = get_filtered_xrefs('biogrid', 'ncbigene')
+
+        biogrid_ncbigene_mapping = get_filtered_xrefs("biogrid", "ncbigene")
     """
     df = get_df()
     df = df.loc[df["IDENTIFIER_TYPE"] == "ENTREZ_GENE", ["BIOGRID_ID", "IDENTIFIER_VALUE"]]

pyobo/sources/ccle.py

@@ -1,10 +1,9 @@
-# -*- coding: utf-8 -*-
-
 """Get the CCLE Cells, provided by cBioPortal."""
 
 import tarfile
+from collections.abc import Iterable
 from pathlib import Path
-from typing import Iterable, Optional
+from typing import Optional
 
 import pandas as pd
 import pystow
@@ -25,7 +24,7 @@ class CCLEGetter(Obo):
 
     ontology = bioregistry_key = PREFIX
 
-    def __post_init__(self):  # noqa: D105
+    def __post_init__(self):
         self.data_version = VERSION
 
     def iter_terms(self, force: bool = False) -> Iterable[Term]:

pyobo/sources/cgnc.py

@@ -1,9 +1,7 @@
-# -*- coding: utf-8 -*-
-
 """Converter for CGNC."""
 
 import logging
-from typing import Iterable
+from collections.abc import Iterable
 
 import pandas as pd
 

pyobo/sources/chebi.py

@@ -1,8 +1,6 @@
-# -*- coding: utf-8 -*-
-
 """Converter for ChEBI."""
 
-from typing import Mapping, Set, Tuple
+from collections.abc import Mapping
 
 from ..api import get_filtered_properties_mapping, get_filtered_relations_df
 from ..struct import Reference, TypeDef
@@ -33,7 +31,7 @@ def get_chebi_smiles_id_mapping() -> Mapping[str, str]:
 has_role = TypeDef(reference=Reference(prefix="chebi", identifier="has_role"))
 
 
-def get_chebi_role_to_children() -> Mapping[str, Set[Tuple[str, str]]]:
+def get_chebi_role_to_children() -> Mapping[str, set[tuple[str, str]]]:
     """Get the ChEBI role to children mapping."""
     df = get_filtered_relations_df("chebi", relation=has_role)
     return multisetdict((role_id, ("chebi", chemical_id)) for chemical_id, _, role_id in df.values)

pyobo/sources/chembl.py

@@ -1,13 +1,11 @@
-# -*- coding: utf-8 -*-
-
 """Converter for ChEMBL.
 
 Run with ``python -m pyobo.sources.chembl -vv``.
 """
 
 import logging
+from collections.abc import Iterable
 from contextlib import closing
-from typing import Iterable
 
 import chembl_downloader
 

pyobo/sources/civic_gene.py

@@ -1,8 +1,7 @@
-# -*- coding: utf-8 -*-
-
 """Converter for CiVIC Genes."""
 
-from typing import Iterable, Optional
+from collections.abc import Iterable
+from typing import Optional
 
 import pandas as pd
 

pyobo/sources/complexportal.py

@@ -1,9 +1,7 @@
-# -*- coding: utf-8 -*-
-
 """Converter for ComplexPortal."""
 
 import logging
-from typing import Iterable, List, Tuple
+from collections.abc import Iterable
 
 import pandas as pd
 from tqdm.auto import tqdm
@@ -52,7 +50,7 @@ DTYPE = {
 }
 
 
-def _parse_members(s) -> List[Tuple[Reference, str]]:
+def _parse_members(s) -> list[tuple[Reference, str]]:
     if pd.isna(s):
         return []
 
@@ -60,15 +58,35 @@ 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
 
 
-def _parse_xrefs(s) -> List[Tuple[Reference, str]]:
+def _parse_xrefs(s) -> list[tuple[Reference, str]]:
     if pd.isna(s):
         return []
 
@@ -76,27 +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)
@@ -105,6 +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/conso.py

@@ -1,8 +1,6 @@
-# -*- coding: utf-8 -*-
-
 """Converter for CONSO."""
 
-from typing import Iterable, List
+from collections.abc import Iterable
 
 import pandas as pd
 
@@ -68,7 +66,7 @@ def iter_terms() -> Iterable[Term]:
     for _, row in terms_df.iterrows():
         if row["Name"] == "WITHDRAWN":
             continue
-        provenance: List[Reference] = []
+        provenance: list[Reference] = []
         for curie in row["References"].split(","):
             curie = curie.strip()
             if not curie:

pyobo/sources/cpt.py

@@ -1,8 +1,6 @@
-# -*- coding: utf-8 -*-
-
 """Converter for CPT."""
 
-from typing import Iterable
+from collections.abc import Iterable
 
 import pandas as pd
 

pyobo/sources/credit.py

@@ -3,7 +3,7 @@
 from __future__ import annotations
 
 import json
-from typing import Iterable
+from collections.abc import Iterable
 
 from more_itertools import chunked
 

pyobo/sources/cvx.py

@@ -1,9 +1,7 @@
-# -*- coding: utf-8 -*-
-
 """Converter for CVX."""
 
 from collections import defaultdict
-from typing import Iterable
+from collections.abc import Iterable
 
 import pandas as pd
 

pyobo/sources/depmap.py

@@ -1,8 +1,7 @@
-# -*- coding: utf-8 -*-
-
 """DepMap cell lines."""
 
-from typing import Iterable, Optional
+from collections.abc import Iterable
+from typing import Optional
 
 import pandas as pd
 import pystow
@@ -113,7 +112,7 @@ def ensure(version: str, force: bool = False) -> pd.DataFrame:
         url=get_url(version=version),
         name="sample_info.tsv",
         force=force,
-        read_csv_kwargs=dict(sep=",", dtype=str),
+        read_csv_kwargs={"sep": ",", "dtype": str},
     )
 
 

pyobo/sources/dictybase_gene.py

@@ -1,18 +1,15 @@
-# -*- coding: utf-8 -*-
-
 """Converter for dictyBase gene.
 
 Note that normal dictybase idenififers are for sequences
 """
 
 import logging
-from typing import Iterable
+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__ = [
@@ -51,10 +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
@@ -70,10 +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/drugbank.py

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 """Convert DrugBank to OBO.
 
 Run with ``python -m pyobo.sources.drugbank``
@@ -8,14 +6,15 @@ Run with ``python -m pyobo.sources.drugbank``
 import datetime
 import itertools as itt
 import logging
+from collections.abc import Iterable, Mapping
 from functools import lru_cache
-from typing import Any, Dict, Iterable, Mapping, Optional
+from typing import Any, Optional
 from xml.etree import ElementTree
 
 import pystow
 from tqdm.auto import tqdm
 
-from ..getters import NoBuild
+from ..getters import NoBuildError
 from ..struct import Obo, Reference, Term
 from ..struct.typedef import has_inchi, has_salt, has_smiles
 from ..utils.cache import cached_pickle
@@ -139,7 +138,7 @@ def _make_term(drug_info: Mapping[str, Any]) -> Term:
     return term
 
 
-@lru_cache()
+@lru_cache
 def get_xml_root(version: Optional[str] = None) -> ElementTree.Element:
     """Get the DrugBank XML parser root.
 
@@ -152,7 +151,7 @@ def get_xml_root(version: Optional[str] = None) -> ElementTree.Element:
         username = pystow.get_config("pyobo", "drugbank_username", raise_on_missing=True)
         password = pystow.get_config("pyobo", "drugbank_password", raise_on_missing=True)
     except ConfigError as e:
-        raise NoBuild from e
+        raise NoBuildError from e
 
     element = parse_drugbank(version=version, username=username, password=password)
     return element.getroot()
@@ -167,7 +166,7 @@ smiles_template = f"{ns}calculated-properties/{ns}property[{ns}kind='SMILES']/{n
 def _extract_drug_info(drug_xml: ElementTree.Element) -> Mapping[str, Any]:
     """Extract information from an XML element representing a drug."""
     # assert drug_xml.tag == f'{ns}drug'
-    row: Dict[str, Any] = {
+    row: dict[str, Any] = {
         "type": drug_xml.get("type"),
         "drugbank_id": drug_xml.findtext(f"{ns}drugbank-id[@primary='true']"),
         "cas": drug_xml.findtext(f"{ns}cas-number"),

pyobo/sources/drugbank_salt.py

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 """Convert DrugBank Salts to OBO.
 
 Run with ``python -m pyobo.sources.drugbank_salt``
@@ -10,11 +8,12 @@ Get relations between drugbank salts and drugbank parents with
 .. code-block:: python
 
     import pyobo
-    df = pyobo.get_filtered_relations_df('drugbank', 'obo:has_salt')
+
+    df = pyobo.get_filtered_relations_df("drugbank", "obo:has_salt")
 """
 
 import logging
-from typing import Iterable
+from collections.abc import Iterable
 
 from .drugbank import iterate_drug_info
 from ..struct import Obo, Reference, Term

pyobo/sources/drugcentral.py

@@ -1,11 +1,9 @@
-# -*- coding: utf-8 -*-
-
 """Get DrugCentral as OBO."""
 
 import logging
 from collections import defaultdict
+from collections.abc import Iterable
 from contextlib import closing
-from typing import DefaultDict, Iterable, List
 
 import bioregistry
 import psycopg2
@@ -25,9 +23,9 @@ PREFIX = "drugcentral"
 HOST = "unmtid-dbs.net"
 PORT = 5433
 USER = "drugman"
-PASSWORD = "dosage"
+PASSWORD = "dosage"  # noqa:S105
 DBNAME = "drugcentral"
-PARAMS = dict(dbname=DBNAME, user=USER, password=PASSWORD, host=HOST, port=PORT)
+PARAMS = {"dbname": DBNAME, "user": USER, "password": PASSWORD, "host": HOST, "port": PORT}
 
 
 class DrugCentralGetter(Obo):
@@ -58,7 +56,7 @@ def iter_terms() -> Iterable[Term]:
         with closing(conn.cursor()) as cur:
             cur.execute("SELECT struct_id, id_type, identifier FROM public.identifier")
             rows = cur.fetchall()
-            xrefs: DefaultDict[str, List[Reference]] = defaultdict(list)
+            xrefs: defaultdict[str, list[Reference]] = defaultdict(list)
             for drugcentral_id, prefix, identifier in tqdm(
                 rows, unit_scale=True, desc="loading xrefs"
             ):
@@ -70,13 +68,16 @@ 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)
                 )
         with closing(conn.cursor()) as cur:
             cur.execute("SELECT id, name FROM public.synonyms")
-            synonyms: DefaultDict[str, List[Synonym]] = defaultdict(list)
+            synonyms: defaultdict[str, list[Synonym]] = defaultdict(list)
             for drugcentral_id, synonym in cur.fetchall():
                 synonyms[str(drugcentral_id)].append(Synonym(name=synonym))
 
@@ -100,4 +101,4 @@ def iter_terms() -> Iterable[Term]:
 
 
 if __name__ == "__main__":
-    get_obo().write_default(write_obo=True)
+    DrugCentralGetter.cli()

pyobo/sources/expasy.py

@@ -1,10 +1,10 @@
-# -*- coding: utf-8 -*-
-
 """Convert ExPASy to OBO."""
 
 import logging
+import re
 from collections import defaultdict
-from typing import Any, Dict, Iterable, Mapping, Optional, Set, Tuple
+from collections.abc import Iterable, Mapping
+from typing import Any, Optional
 
 from .utils import get_go_mapping
 from ..struct import Obo, Reference, Synonym, Term
@@ -43,7 +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"),
@@ -76,7 +76,7 @@ def get_terms(version: str, force: bool = False) -> Iterable[Term]:
     with open(tree_path) as file:
         tree = get_tree(file)
 
-    terms: Dict[str, Term] = {}
+    terms: dict[str, Term] = {}
     child_to_parents = defaultdict(list)
     for ec_code, data in tree.items():
         terms[ec_code] = Term(
@@ -146,7 +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(
@@ -176,7 +178,7 @@ def normalize_expasy_id(expasy_id: str) -> str:
     return expasy_id.replace(" ", "")
 
 
-def give_edge(unnormalized_ec_code: str) -> Tuple[int, Optional[str], str]:
+def give_edge(unnormalized_ec_code: str) -> tuple[int, Optional[str], str]:
     """Return a (parent, child) tuple for given id."""
     levels = [x for x in unnormalized_ec_code.replace(" ", "").replace("-", "").split(".") if x]
     level = len(levels)
@@ -227,7 +229,7 @@ def get_database(lines: Iterable[str]) -> Mapping:
     for groups in _group_by_id(lines):
         _, expasy_id = groups[0]
 
-        ec_data_entry: Dict[str, Any] = {
+        ec_data_entry: dict[str, Any] = {
             "concept": {
                 "namespace": PREFIX,
                 "identifier": expasy_id,
@@ -249,8 +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:
@@ -269,17 +273,30 @@ def get_database(lines: Iterable[str]) -> Mapping:
                         continue
                     uniprot_id, uniprot_accession = uniprot_entry.split(",")
                     ec_data_entry["proteins"].append(  # type:ignore
-                        dict(
-                            namespace="uniprot",
-                            name=uniprot_accession,
-                            identifier=uniprot_id,
-                        )
+                        {
+                            "namespace": "uniprot",
+                            "name": uniprot_accession,
+                            "identifier": uniprot_id,
+                        }
                     )
 
         rv[expasy_id] = ec_data_entry
     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 = []
@@ -300,7 +317,7 @@ def _group_by_id(lines):
     return groups
 
 
-def get_ec2go(version: str) -> Mapping[str, Set[Tuple[str, str]]]:
+def get_ec2go(version: str) -> Mapping[str, set[tuple[str, str]]]:
     """Get the EC mapping to GO activities."""
     url = "http://current.geneontology.org/ontology/external2go/ec2go"
     path = ensure_path(PREFIX, url=url, name="ec2go.tsv", version=version)