CogniChem's Toolkit

Leverage our message-passing neural networks for accurate and interpretable structure-property modeling and lightning fast inference. Select from our models or train your own to fit your specific research needs.

Boltz-2 is a next-generation biomolecular model for structure and binding prediction. It matches the accuracy of physics-based FEP methods at 1000x the speed, making rapid, accurate in silico screening possible for early-stage drug discovery.

BoltzGen is an all-atom generative model for designing proteins and peptides to bind diverse biomolecular targets. It unifies design and structure prediction, enabling state-of-the-art folding and controllable generation for a wide range of applications.

Compute thousands of molecular descriptors and fingerprints using PaDEL-Descriptor. Enable QSAR modeling, clustering, and chemical informatics with robust descriptor calculation.

Seamlessly convert between chemical formats: SMILES, InChI, PDB, MOL, and more. Integrate with your workflow and ensure compatibility across cheminformatics platforms.