Improving de novo protein binder design with deep learning.

Journal: Nature communications
Published Date:

Abstract

Recently it has become possible to de novo design high affinity protein binding proteins from target structural information alone. There is, however, considerable room for improvement as the overall design success rate is low. Here, we explore the augmentation of energy-based protein binder design using deep learning. We find that using AlphaFold2 or RoseTTAFold to assess the probability that a designed sequence adopts the designed monomer structure, and the probability that this structure binds the target as designed, increases design success rates nearly 10-fold. We find further that sequence design using ProteinMPNN rather than Rosetta considerably increases computational efficiency.

Authors

  • Nathaniel R Bennett
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Brian Coventry
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Inna Goreshnik
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Buwei Huang
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Aza Allen
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Dionne Vafeados
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Ying Po Peng
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Justas Dauparas
    Department of Biochemistry and Institute for Protein Design, University of Washington, Washington, WA, USA.
  • Minkyung Baek
    Department of Biochemistry and Institute for Protein Design, University of Washington, Washington, WA, USA.
  • Lance Stewart
    Department of Biochemistry, University of Washington, Seattle, WA, USA.
  • Frank DiMaio
    Department of Biochemistry, University of Washington, Seattle, WA 98195, USA.
  • Steven De Munck
    VIB-UGent Center for Inflammation Research, Ghent, Belgium.
  • Savvas N Savvides
    VIB-UGent Center for Inflammation Research, Ghent, Belgium.
  • David Baker
    Department of Biochemistry, University of Washington, Seattle, Washington.