De novo design of protein structure and function with RFdiffusion

TLDR

Recent deep‑learning advances have enabled new protein designs, yet a unified framework capable of tackling diverse challenges such as binder, symmetric oligomer, and enzyme scaffold design remains lacking, partly because diffusion models have struggled with protein backbone complexity. This work aims to create a general deep‑learning framework that can generate protein backbones for a wide range of design tasks by fine‑tuning RoseTTAFold on denoising objectives. The authors fine‑tune the RoseTTAFold structure prediction network on protein structure denoising tasks to produce a diffusion‑based generative model, RFdiffusion, which can design unconditional and topology‑constrained monomers, binders, symmetric oligomers, enzyme active sites, and motif scaffolds. RFdiffusion achieves outstanding performance across these tasks, with experimental validation of hundreds of designed symmetric assemblies, metal‑binding proteins, and binders, and cryo‑EM confirmation that a binder’s structure matches the design model nearly perfectly, demonstrating the method’s accuracy and versatility.

Abstract

Abstract There has been considerable recent progress in designing new proteins using deep-learning methods 1–9 . Despite this progress, a general deep-learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher-order symmetric architectures, has yet to be described. Diffusion models 10,11 have had considerable success in image and language generative modelling but limited success when applied to protein modelling, probably due to the complexity of protein backbone geometry and sequence–structure relationships. Here we show that by fine-tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of designed symmetric assemblies, metal-binding proteins and protein binders. The accuracy of RFdiffusion is confirmed by the cryogenic electron microscopy structure of a designed binder in complex with influenza haemagglutinin that is nearly identical to the design model. In a manner analogous to networks that produce images from user-specified inputs, RFdiffusion enables the design of diverse functional proteins from simple molecular specifications.