Description:
Uni-Mol is a 3D molecular pretraining framework that has been trained on over 209 million molecular conformations and 3 million protein pockets. It outperforms many existing molecular modeling methods in predicting molecular properties, protein-ligand interactions, and molecular docking. Researchers can use Uni-Mol for drug discovery, material science, and bioinformatics.
Topic Creator Motivation:
Discuss the applications of Uni-Mol in drug discovery and material science. Join to share methodologies, dataset challenges, and troubleshooting tips.