embl-ebi-ols

Run search_ols.py with --query and optionally --ontology to filter results, for example --query "diabetes" --ontology doid. Results are written as JSON to the file given by --output, including labels, OBO IDs, descriptions, and synonyms.

Use get_term.py with --obo_id and --relations parents,children to fetch direct is-a relations. For transitive relations like part-of or develops-from, use hierarchicalParents or hierarchicalChildren instead.

Use doid for common human diseases, hp for phenotypes, go for gene functions, chebi for chemicals, uberon for anatomy, and cl for cell types. Use mondo only when cross-species disease context is explicitly needed.

Yes, EBI's Terms of Use allow a maximum of 5 requests per second. The provided scripts enforce this automatically through a shared HTTP client configured at 5 queries per second, so no manual throttling is needed.

Use search_ols.py with --exact and --ontology to perform exact label matching against a single ontology, then read the obo_id field from the JSON output. This two-step pattern resolves a name before fetching its relations.

Search defaults to 10 results per page with a maximum of 500 rows, so large result sets require pagination via --start and --rows. Obsolete terms are excluded unless --obsolete is passed, and cross-ontology searches can return noisy duplicates without --defining.