Relay alphafold! Star pharmaceutical science and technology released a new era of tbind opening molecular protein complex structure prediction

Following 2018 year 、2020 The two generation of the year AlphaFold After the birth of , Protein structure prediction is close to experimental accuracy , It has laid a solid foundation for the research and development of computational enabling drugs . In the long process of drug research and development , Molecular protein interaction prediction is the next step in the field of computational enabled drug development “ grail ”. In recent days, , Star pharmaceutical technology pre released Trigonometry Aware Neural NetworK for Drug-Protein Binding Structure Prediction. This model is the first in-depth representation learning framework that can simultaneously predict the three-dimensional binding conformation and binding affinity of small molecules and target proteins , The best results that go well beyond existing methods .TBind Adopt an end-to-end data-driven paradigm , Geometric neural network combined with physical heuristics , The double prediction of the three-dimensional bonding mode and bonding strength of the composite has been achieved , It has achieved the double transcendence of the accuracy and efficiency of the international commercial molecular docking software , It provides the first breakthrough scheme for molecular protein interaction prediction in China . Following AlphaFold after , Bring computing driven drug development into a new era .

Paper title ：

TANKBind: Trigonometry-Aware Neural NetworKs for Drug-Protein Binding Structure Prediction

Thesis link ：

https://www.biorxiv.org/content/biorxiv/early/2022/06/06/2022.06.06.495043.full.pdf

TBind Schematic diagram of model . The input of the model is a three-dimensional protein structure and a three-dimensional molecular structure , The output is the combination mode and strength between the two .

At present, the research team is GitHub Released a free open source beta version TBind v0.5.0, And provide case demonstration .

Open source links ：

https://github.com/luwei0917/TankBind