| Literature DB >> 36108050 |
J Dauparas1,2, I Anishchenko1,2, N Bennett1,2,3, H Bai1,2,4, R J Ragotte1,2, L F Milles1,2, B I M Wicky1,2, A Courbet1,2,4, R J de Haas5, N Bethel1,2,4, P J Y Leung1,2,3, T F Huddy1,2, S Pellock1,2, D Tischer1,2, F Chan1,2, B Koepnick1,2, H Nguyen1,2, A Kang1,2, B Sankaran6, A K Bera1,2, N P King1,2, D Baker1,2,4.
Abstract
Although deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here, we describe a deep learning-based protein sequence design method, ProteinMPNN, that has outstanding performance in both in silico and experimental tests. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4% compared with 32.9% for Rosetta. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. We demonstrate the broad utility and high accuracy of ProteinMPNN using x-ray crystallography, cryo-electron microscopy, and functional studies by rescuing previously failed designs, which were made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target-binding proteins.Entities:
Mesh：
Substances：
Year: 2022 PMID： 36108050 DOI： 10.1126/science.add2187
Source DB: PubMed Journal: Science ISSN： 0036-8075 Impact factor: 63.714