4 datasets found

Deep generative models have emerged as a powerful tool for learning useful molecular representations and designing novel molecules with desired properties, with applications in...

The dataset used in this study consists of 1.6M chemical molecules from the MOSES benchmarking dataset, 211k ligand molecules from the BindingDB database, and 12 in vitro...

The SELFIES dataset is a string representation of molecules, which is used as input to the proposed CHA2 method for inverse molecular design.

The ZINC250K dataset is a large dataset of molecules used for molecular design and generation. It contains 250,000 molecules with their corresponding properties and structures.