De novo design of high-affinity protein binders with AlphaProteo

• DOI: 10.48550/arxiv.2409.08022
• Publication Date: 2024-09-12
• Authors: Vinícius Zambaldi; David La; Alexander E. Chu; Harshnira Patani; Amy E. Danson; Tristan O. C. Kwan; Thomas Frerix; Rosalia G. Schneider; David Saxton; Ashok Thillaisund...; Zachary Wu; Isabel Moraes; O. Lange; Eliseo Papa; Gabriella Stanton; Victor S. Martı́n; Sukhdeep Singh; Lai Hong Wong; Russ Bates; Simon Kohl; Josh Abramson; Andrew Senior; Yilmaz Alguel; Mary Wu; Irene M. Aspalter; Katie Bentley; David L.V. Bauer; Peter Cherepanov; Demis Hassabis; Pushmeet Kohli; Rob Fergus; Jue Wang
• Keywords: Quantitative Biology - Biomolecules; FOS: Biological sciences; Biomolecules (q-bio.BM)

Computational design of protein-binding proteins is a fundamental capability with broad utility in biomedical research and biotechnology. Recent methods have made strides against some target proteins, but on-demand creation high-affinity binders without multiple rounds experimental testing remains an unsolved challenge. This technical report introduces AlphaProteo, family machine learning models for protein design, details its performance on the de novo binder problem. With we achieve 3- to 300-fold better binding affinities higher success rates than best existing seven proteins. Our results suggest that AlphaProteo can generate "ready-to-use" many applications using only one round medium-throughput screening no further optimization.