DynamicBind is a deep learning method using equivariant geometric diffusion networks to facilitate efficient transitions between protein equilibrium states. It recovers ligand-specific conformations from unbound protein structures promoting efficient transitions between different equilibrium states.

TankBind is a geometric deep learning method for molecular rigid docking. It could predict both the protein-ligand binding structure and their affinity. And it also support virtual screening.

BioNavi-NP is a user-friendly toolkit that predicts biosynthetic pathways for NPs and NP-like compounds.The toolkit as well as the curated datasets and learned models are freely available to facilitate the elucidation and reconstruction of the biosynthetic pathways for NPs.

MolRep is a Python package for fairly measuring algorithmic progress on chemical property datasets. It currently provides a complete re-evaluation of 16 state-of-the-art deep representation models over 16 benchmark property datsaets.

DrugVQA is a multimodel learning method for protein-ligand interaction prediction.