RDKit Cheminformatics Toolkit

RDKit Cheminformatics Toolkit

Overview

RDKit is the standard open-source cheminformatics library for Python, providing comprehensive APIs for molecular parsing, descriptor calculation, fingerprinting, substructure searching, and chemical reactions. This skill walks through a complete compound library profiling and virtual screening workflow — from loading molecules through drug-likeness filtering, similarity screening, and result visualization.

When to Use

• Calculate molecular properties (MW, LogP, TPSA, HBD/HBA) for a compound set
• Screen a library against a reference compound using fingerprint similarity
• Filter compounds by substructure (SMARTS patterns) for functional group analysis
• Assess drug-likeness using Lipinski's Rule of Five or custom filters
• Generate 2D depictions or 3D conformers for downstream docking
• Enumerate chemical libraries using reaction SMARTS (combinatorial chemistry)
• Cluster compounds by structural similarity for diversity analysis
• Standardize and deduplicate molecular datasets (canonical SMILES, InChI)
• Use datamol-cheminformatics instead for a higher-level RDKit wrapper with batching and error handling; use openbabel instead for multi-format conversion (MOL2, XYZ, PDB)

Prerequisites

• Python packages: rdkit-pypi (or rdkit via conda), pandas, matplotlib, numpy
• Data requirements: Molecular structures as SMILES strings, SDF files, or MOL files
• Environment: Python 3.8+; conda recommended for full RDKit installation

# Option 1: pip (lightweight)
pip install rdkit-pypi pandas matplotlib numpy

# Option 2: conda (full features including cartridge)
conda install -c conda-forge rdkit pandas matplotlib numpy

Workflow

Step 1: Load and Validate Molecules

Read molecular structures from SMILES or SDF and validate parsing.

from rdkit import Chem