Molclaw Drug Likeness

Compute the drug-likeness metrics (QED score and Number of violations of Lipinski's Rule of Five) of the input candidate molecules (SMILES format).

internscience Updated 607 repo stars
Agents-A1 Benchmark Overview

File contents

Molecular Drug-likeness Metrics Calculation

The description of tool calculate_mol_drug_chemistry.

Compute key drug-likeness metrics for each SMILES.
Args:
    smiles_list (List[str]): List of input SMILES strings, (e.g., ["N[C@@H](Cc1ccc(O)cc1)C(=O)O", "CC(C)C1=CC=CC=C1"])
Return:
    status (str): success/error
    msg (str): message
    metrics (List[dict]): List of dict, each containing feature keys.
        --smiles (str): A SMILES string of smiles_list
        --qed (float): Quantitative Estimate of Drug-likeness (QED) score
        --lipinski_rule_of_5_violations (int): Number of violations of Lipinski's Rule of Five

How to use tool calculate_mol_drug_chemistry :

response = await client.session.call_tool(
    "calculate_mol_drug_chemistry",
    arguments={
        "smiles_list": smiles_list
    }
)
result = client.parse_result(response)
druglikeness_metrics = result["metrics"]

internscience/agents-a1 commit 4ff99b80c9

Frequently asked questions

npx get-skill add internscience/molclaw-drug-likeness-2