generate_wikipathways_orthologs.py

"""Generate orthologous relations between WikiPathways."""

import itertools as itt
from collections.abc import Iterable
from typing import cast

import pyobo
from bioregistry import NormalizedNamableReference
from curies.vocabulary import lexical_matching_process
from gilda.process import normalize
from sssom_pydantic import MappingTool, SemanticMapping
from tqdm import tqdm

from biomappings.resources import append_predictions
from biomappings.utils import get_script_url


def _lexical_exact_match(name1: str, name2: str) -> bool:
    return cast(str, normalize(name1)) == cast(str, normalize(name2))  # type:ignore[no-untyped-call]


def iterate_orthologous_lexical_matches(prefix: str = "wikipathways") -> Iterable[SemanticMapping]:
    """Generate orthologous relations between lexical matches from different species."""
    names = pyobo.get_id_name_mapping(prefix)
    species = pyobo.get_id_species_mapping(prefix)
    provenance = get_script_url(__file__)

    count = 0
    it = itt.combinations(sorted(names.items()), 2)
    it = tqdm(
        it,
        unit_scale=True,
        unit="pair",
        total=len(names) * (len(names) - 1) / 2,
    )
    for (source_id, source_name), (target_id, target_name) in sorted(it):
        source_species = species[source_id]
        target_species = species[target_id]
        if source_species == target_species:
            continue
        if _lexical_exact_match(source_name, target_name):
            count += 1
            yield SemanticMapping(
                subject=NormalizedNamableReference(
                    prefix=prefix, identifier=source_id, name=source_name
                ),
                predicate="RO:HOM0000017",
                object=NormalizedNamableReference(
                    prefix=prefix,
                    identifier=target_id,
                    name=target_name,
                ),
                justification=lexical_matching_process,
                confidence=0.95,
                mapping_tool=MappingTool(name=provenance),
            )
    tqdm.write(f"Identified {count:,} orthologs")


if __name__ == "__main__":
    append_predictions(iterate_orthologous_lexical_matches())