rdkit-chemistry

RDKit Chemistry

Molecular chemistry operations using RDKit.

When to Use

• Molecular structures, SMILES parsing, or validation
• Molecular properties (MW, logP, TPSA, HBD/HBA)
• Substructure searching or molecular filtering
• Fingerprints (Morgan, MACCS) and similarity calculations
• 2D depiction or molecular image generation

When NOT to Use

• Reaction databases or retrosynthesis planning
• Wet lab protocols or experimental procedures
• Protein structure analysis (use biopython-bio)
• Quantum chemistry or DFT calculations

SMILES Parsing and Properties

from rdkit import Chem
from rdkit.Chem import Descriptors, rdMolDescriptors

mol = Chem.MolFromSmiles('CC(=O)Oc1ccccc1C(=O)O')  # Aspirin
if mol is None:
    print("Invalid SMILES")

canonical = Chem.MolToSmiles(mol)                      # Canonical SMILES
mw = Descriptors.MolWt(mol)                            # Molecular weight
logp = Descriptors.MolLogP(mol)                        # Partition coefficient
tpsa = Descriptors.TPSA(mol)                           # Topological polar surface area
hbd = rdMolDescriptors.CalcNumHBD(mol)                 # H-bond donors
hba = rdMolDescriptors.CalcNumHBA(mol)                 # H-bond acceptors
rotatable = rdMolDescriptors.CalcNumRotatableBonds(mol)

# SMARTS substructure match
pattern = Chem.MolFromSmarts('[OH]')
has_oh = mol.HasSubstructMatch(pattern)

Lipinski Rule of Five

def lipinski(smi):
    mol = Chem.MolFromSmiles(smi)
    return {
        'MW <= 500': Descriptors.MolWt(mol) <= 500,
        'LogP <= 5': Descriptors.MolLogP(mol) <= 5,
        'HBD <= 5': rdMolDescriptors.CalcNumHBD(mol) <= 5,
        'HBA <= 10': rdMolDescriptors.CalcNumHBA(mol) <= 10,
    }

Substructure Search

molecules = [Chem.MolFromSmiles(s) for s in ['CCO', 'CC(=O)O', 'c1ccccc1', 'c1ccccc1O']]
pattern = Chem.MolFromSmarts('c1ccccc1')  # Benzene ring
hits = [m for m in molecules if m.HasSubstructMatch(pattern)]

Fingerprints and Similarity