Robust deep learning–based protein sequence design using ProteinMPNN

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…

• NS
  National Science Foundation

  Awards: DGE-2140004, DBI 1937533, 1937533, DE-AC02-05CH11231, P30 GM124169, 2140004
• UD
  U.S. Department of Energy

  Awards: -AC02-05CH11231, 05CH11231, AC02-05CH11231, DE-AC02, DE-AC02-05CH11231, GM124169, DE-AC02-
• AP
  Alfred P. Sloan Foundation

  Awards: G-2021-16899, DE-AC02-05CH11231
• WR
  Washington Research Foundation
• OP
  Open Philanthropy Project
• NI
  National Institutes of Health

  Awards: GM124169, P30 GM124169-01, P30 GM124169, DE-AC02-05CH11231
• OO
  Office of Science

  Awards: AC02-05CH11231, -AC02-05CH11231, DE-AC02
• NI
  National Institute of General Medical Sciences

  Awards: DE-AC02-05CH11231, P30 GM124169, P30 GM124169-01, GM124169-01, GM124169
• BE
  Basic Energy Sciences

  Awards: DE-AC02, AC02-05CH11231, DE-AC02-05CH11231, -AC02-05CH11231