Update genesis/api_clients/chembl_api.py

genesis/api_clients/chembl_api.py

@@ -3,69 +3,104 @@ import os
 import requests
 from typing import Dict, List, Optional
 
-CHEMBL_BASE = "https://www.ebi.ac.uk/chembl/api/data"
-CHEMBL_API_KEY = os.getenv("CHEMBL_API_KEY")  # If you have API key access, else can be None (public API works without auth)
+CHEMBL_BASE = "https://www.ebi.ac.uk/chembl/api/data/"
+CHEMBL_API_KEY = os.getenv("CHEMBL_API_KEY")  # Optional if you have API key
 
-def search_molecule(query: str, max_results: int = 10) -> List[Dict]:
+# -------------------------
+# Core Request Helper
+# -------------------------
+def chembl_request(endpoint: str, params: Dict) -> Dict:
     """
-    Search ChEMBL for a molecule by name, synonym, or ChEMBL ID.
+    Helper to make ChEMBL API requests.
     """
-    params = {"format": "json", "limit": max_results, "molecule_synonyms__icontains": query}
-    r = requests.get(f"{CHEMBL_BASE}/molecule", params=params)
+    url = f"{CHEMBL_BASE}{endpoint}"
+    headers = {"Accept": "application/json"}
+    if CHEMBL_API_KEY:
+        headers["Authorization"] = f"Bearer {CHEMBL_API_KEY}"
+    r = requests.get(url, headers=headers, params=params)
     r.raise_for_status()
-    data = r.json()
-    return data.get("molecules", [])
+    return r.json()
 
+# -------------------------
+# Search Molecules
+# -------------------------
+def search_molecule(query: str, max_results: int = 10) -> List[Dict]:
+    """
+    Search for molecules in ChEMBL by name or ChEMBL ID.
+    """
+    data = chembl_request("molecule/search", {"q": query, "limit": max_results})
+    return data.get("molecules", [])
 
 def get_molecule_details(chembl_id: str) -> Dict:
     """
-    Retrieve full details for a specific molecule (properties, SMILES, InChI).
+    Get detailed molecule info by ChEMBL ID.
     """
-    r = requests.get(f"{CHEMBL_BASE}/molecule/{chembl_id}.json")
-    r.raise_for_status()
-    return r.json()
+    return chembl_request(f"molecule/{chembl_id}", {})
 
-
-def get_target_details(target_chembl_id: str) -> Dict:
+def get_molecule_image(chembl_id: str) -> str:
     """
-    Retrieve details about a protein target.
+    Get image (SVG) of a molecule structure.
     """
-    r = requests.get(f"{CHEMBL_BASE}/target/{target_chembl_id}.json")
-    r.raise_for_status()
-    return r.json()
-
+    return f"https://www.ebi.ac.uk/chembl/api/data/image/{chembl_id}.svg"
 
-def get_bioactivities(chembl_id: str, max_results: int = 20) -> List[Dict]:
+# -------------------------
+# Targets
+# -------------------------
+def search_target(query: str, max_results: int = 10) -> List[Dict]:
     """
-    Retrieve bioactivity assays for a given molecule.
+    Search for biological targets in ChEMBL.
     """
-    params = {"format": "json", "limit": max_results}
-    r = requests.get(f"{CHEMBL_BASE}/activity?molecule_chembl_id={chembl_id}", params=params)
-    r.raise_for_status()
-    return r.json().get("activities", [])
+    data = chembl_request("target/search", {"q": query, "limit": max_results})
+    return data.get("targets", [])
 
-
-def get_mechanism_of_action(chembl_id: str) -> List[Dict]:
+def get_target_details(chembl_id: str) -> Dict:
     """
-    Retrieve mechanism of action data for a drug.
+    Get detailed target info from ChEMBL.
     """
-    r = requests.get(f"{CHEMBL_BASE}/mechanism.json?molecule_chembl_id={chembl_id}")
-    r.raise_for_status()
-    return r.json().get("mechanisms", [])
+    return chembl_request(f"target/{chembl_id}", {})
 
+# -------------------------
+# Bioactivity
+# -------------------------
+def get_bioactivity_for_molecule(chembl_id: str, max_results: int = 20) -> List[Dict]:
+    """
+    Get bioactivity data for a molecule.
+    """
+    data = chembl_request("activity", {"molecule_chembl_id": chembl_id, "limit": max_results})
+    return data.get("activities", [])
 
-def get_clinical_trials(chembl_id: str) -> List[Dict]:
+def get_bioactivity_for_target(chembl_id: str, max_results: int = 20) -> List[Dict]:
     """
-    Retrieve clinical trial data linked to a drug.
+    Get bioactivity data for a biological target.
     """
-    r = requests.get(f"{CHEMBL_BASE}/clinical_trial.json?molecule_chembl_id={chembl_id}")
-    r.raise_for_status()
-    return r.json().get("clinical_trials", [])
+    data = chembl_request("activity", {"target_chembl_id": chembl_id, "limit": max_results})
+    return data.get("activities", [])
 
+# -------------------------
+# Assays
+# -------------------------
+def get_assays_for_molecule(chembl_id: str, max_results: int = 20) -> List[Dict]:
+    """
+    Get assay data for a molecule.
+    """
+    data = chembl_request("assay", {"molecule_chembl_id": chembl_id, "limit": max_results})
+    return data.get("assays", [])
 
-def get_structural_image(chembl_id: str, size: str = "500") -> Optional[str]:
+# -------------------------
+# Integration Helpers
+# -------------------------
+def molecule_to_gene_links(chembl_id: str) -> List[Dict]:
     """
-    Retrieve PNG image of molecule from ChEMBL's structure rendering service.
-    Returns image URL.
+    Crosslink molecule targets to gene symbols for integration with NCBI.
     """
-    return f"https://www.ebi.ac.uk/chembl/api/utils/image/{chembl_id}?dimensions={size}"
+    targets = get_bioactivity_for_molecule(chembl_id)
+    gene_links = []
+    for t in targets:
+        if "target_chembl_id" in t:
+            target_info = get_target_details(t["target_chembl_id"])
+            gene_links.append({
+                "target_id": t["target_chembl_id"],
+                "gene_symbol": target_info.get("target_components", [{}])[0].get("accession", ""),
+                "organism": target_info.get("organism", "")
+            })
+    return gene_links