Descriptors & Property Checks

Descriptors let you profile a molecule's physicochemical properties — molecular weight, logP, TPSA, hydrogen-bond donors and acceptors — and gate compounds through drug-likeness checks before expensive steps. The property bounds you set in your QC gate are what most affect how much of your set passes through to conformer generation or docking.

Common descriptors

• Molecular weight: Descriptors.MolWt(mol)
• logP: Descriptors.MolLogP(mol)
• TPSA: Descriptors.TPSA(mol)
• HBA/HBD: Lipinski.NumHAcceptors, Lipinski.NumHDonors
• Rotatable bonds: Lipinski.NumRotatableBonds

from rdkit import Chem
from rdkit.Chem import Descriptors, Lipinski

mol = Chem.MolFromSmiles("CC(=O)Oc1ccccc1C(=O)O")

summary = {
    "mw": Descriptors.MolWt(mol),
    "logp": Descriptors.MolLogP(mol),
    "tpsa": Descriptors.TPSA(mol),
    "hba": Lipinski.NumHAcceptors(mol),
    "hbd": Lipinski.NumHDonors(mol),
    "rot_bonds": Lipinski.NumRotatableBonds(mol),
}
print(summary)

QC gates

def passes_qc(mol) -> bool:
    if mol is None:
        return False
    mw = Descriptors.MolWt(mol)
    logp = Descriptors.MolLogP(mol)
    rb = Lipinski.NumRotatableBonds(mol)
    if not (150 <= mw <= 600):
        return False
    if not (-1.5 <= logp <= 5.5):
        return False
    if rb > 10:
        return False
    return True

Apply passes_qc before expensive steps (conformer generation, docking) to save time.

Batch workflow

• Parse SMILES → run QC → write passing molecules to CSV/SDF.
• Log counts for pass/fail buckets to spot over-filtering.
• For large sets, run QC in chunks and stream outputs to avoid high memory use.