medsci-agent: Skill for OpenCode

ChEMBL Database

Overview

ChEMBL is a large-scale bioactivity database maintained by EMBL-EBI. It contains binding, functional, and ADMET data for drug-like molecules against biological targets.

Python Client

from chembl_webresource_client.new_client import new_client

molecule = new_client.molecule
activity = new_client.activity
target = new_client.target

Search target by name

results = target.search("cyclooxygenase-2")
for t in results:
    print(t["target_chembl_id"], t["pref_name"], t["organism"])

Get bioactivity data for a target

acts = activity.filter(target_chembl_id="CHEMBL220", standard_type="IC50", pchembl_value__isnull=False)
for a in acts:
    print(a["molecule_chembl_id"], a["pchembl_value"], a["canonical_smiles"])

Retrieve molecule by SMILES or ChEMBL ID

mol = molecule.get("CHEMBL25")  # aspirin
mol = molecule.filter(molecule_structures__canonical_smiles="CC(=O)Oc1ccccc1C(=O)O")

Key Fields

• pchembl_value: Standardized -log10(IC50/Ki/EC50) in molar. Use this for comparisons.
• standard_type: IC50, Ki, EC50, Kd, etc.
• assay_type: B (binding), F (functional), A (ADMET).
• target_type: SINGLE PROTEIN, PROTEIN COMPLEX, ORGANISM, etc.

Key Details

• Always filter on pchembl_value__isnull=False for comparable potency data.
• ChEMBL IDs: molecules (CHEMBL25), targets (CHEMBL220), assays (CHEMBL123456).
• Install: pip install chembl-webresource-client.