mcp/ncbi.py

# mcp/ncbi.py
"""
NCBI E-utilities helpers – Gene, Protein, ClinVar, MeSH.
"""

import os
import httpx
from typing import List, Dict

NCBI_KEY = os.getenv("BIO_KEY")  # optional but increases rate limits
BASE = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/"

async def _get(endpoint: str, params: Dict) -> Dict:
    if NCBI_KEY:
        params["api_key"] = NCBI_KEY
    async with httpx.AsyncClient(timeout=20) as client:
        r = await client.get(f"{BASE}{endpoint}", params=params)
        r.raise_for_status()
        return r.json() if r.headers["Content-Type"].startswith("application/json") else r.text

# ---------- Public helpers ----------
async def search_gene(term: str, retmax: int = 5) -> List[Dict]:
    """Return basic gene info (ID + name/symbol) by search term."""
    data = await _get("esearch.fcgi", {"db": "gene", "term": term, "retmode": "json", "retmax": retmax})
    ids = data["esearchresult"]["idlist"]
    if not ids:
        return []
    summary = await _get("esummary.fcgi", {"db": "gene", "id": ",".join(ids), "retmode": "json"})
    return list(summary["result"].values())[1:]  # first key is 'uids'

async def get_mesh_definition(term: str) -> str:
    """Return MeSH term definition (first record)."""
    text = await _get("esummary.fcgi", {"db": "mesh", "term": term, "retmode": "json", "retmax": 1})
    recs = list(text["result"].values())[1:]
    return recs[0].get("ds_meshterms", [""])[0] if recs else ""