genesis/api_clients/chembl_api.py

# genesis/api_clients/chembl_api.py
import os
import requests
from typing import List, Dict, Optional

CHEMBL_BASE = "https://www.ebi.ac.uk/chembl/api/data"

def _fetch_paginated(endpoint: str, params: dict) -> List[dict]:
    """
    Handles paginated ChEMBL API responses.
    """
    results = []
    url = f"{CHEMBL_BASE}/{endpoint}"
    
    while url:
        r = requests.get(url, params=params)
        r.raise_for_status()
        data = r.json()
        page_items = data.get(endpoint, [])
        results.extend(page_items)
        url = data.get("page_meta", {}).get("next")
        params = {}  # clear params after first page to avoid duplication
    return results


def search_molecule_by_name(name: str, max_results: int = 10) -> List[dict]:
    """
    Search molecules in ChEMBL by name or synonym.
    """
    params = {"molecule_synonyms__icontains": name, "limit": max_results}
    return _fetch_paginated("molecule", params)


def get_molecule_details(chembl_id: str) -> dict:
    """
    Retrieve detailed information for a molecule by ChEMBL ID.
    """
    r = requests.get(f"{CHEMBL_BASE}/molecule/{chembl_id}")
    r.raise_for_status()
    return r.json()


def get_molecule_bioactivities(chembl_id: str, max_results: int = 50) -> List[dict]:
    """
    Retrieve bioactivity records for a given molecule.
    """
    params = {"molecule_chembl_id": chembl_id, "limit": max_results}
    return _fetch_paginated("activity", params)


def search_target_by_name(name: str, max_results: int = 10) -> List[dict]:
    """
    Search for biological targets by name.
    """
    params = {"target_synonym__icontains": name, "limit": max_results}
    return _fetch_paginated("target", params)


def get_target_details(chembl_id: str) -> dict:
    """
    Retrieve detailed target information by ChEMBL ID.
    """
    r = requests.get(f"{CHEMBL_BASE}/target/{chembl_id}")
    r.raise_for_status()
    return r.json()


def search_assays_by_target(target_chembl_id: str, max_results: int = 50) -> List[dict]:
    """
    Retrieve assays related to a specific target.
    """
    params = {"target_chembl_id": target_chembl_id, "limit": max_results}
    return _fetch_paginated("assay", params)


def get_drugs_in_clinical_phase(phase: int, max_results: int = 20) -> List[dict]:
    """
    Retrieve drugs by clinical trial phase.
    """
    params = {"max_phase": phase, "limit": max_results}
    return _fetch_paginated("molecule", params)


def get_molecule_image_url(chembl_id: str) -> str:
    """
    Generate a URL to the 2D structure image of a molecule.
    """
    return f"https://www.ebi.ac.uk/chembl/api/data/image/{chembl_id}.svg"