Simplifies molecular cheminformatics and drug discovery workflows using a Pythonic interface for RDKit.
Datamol is a lightweight abstraction layer over RDKit designed to make molecular operations more accessible and efficient for researchers and developers. It provides a simplified interface for common drug discovery tasks such as structure standardization, descriptor calculation, fingerprinting, and 3D conformer generation. By offering sensible defaults and built-in parallelization, Datamol streamlines the preparation of chemical datasets for machine learning and complex SAR analysis while maintaining full compatibility with native RDKit objects.
Key Features
01Parallelized molecular descriptor and fingerprint computation
02Integrated I/O support for SDF, SMILES, and cloud storage (S3/GCS)
03Efficient 3D conformer generation and energy minimization
04Advanced scaffold analysis and Bemis-Murcko decomposition
05Molecular structure standardization and sanitization
060 GitHub stars
Use Cases
01Automating molecular property prediction and drug-likeness filtering
02Performing SAR analysis through scaffold-based clustering and grouping
03Preparing large-scale molecular datasets for machine learning models
