ProteinWeaver: A Divide-and-Assembly Approach for Protein Backbone Design

TL;DR

ProteinWeaver introduces a novel two-stage framework inspired by nature's divide-and-assemble strategy, enabling the design of diverse, high-quality protein backbones through domain generation and flexible assembly using diffusion models.

Contribution

It presents a new divide-and-assembly approach for protein backbone design, outperforming existing methods and demonstrating potential for functional protein engineering.

Findings

Outperforms RFdiffusion by 13% and 39% for long-chain proteins

Generates high-quality, novel protein backbones

Shows potential for cooperative function design

Abstract

Nature creates diverse proteins through a 'divide and assembly' strategy. Inspired by this idea, we introduce ProteinWeaver, a two-stage framework for protein backbone design. Our method first generates individual protein domains and then employs an SE(3) diffusion model to flexibly assemble these domains. A key challenge lies in the assembling step, given the complex and rugged nature of the inter-domain interaction landscape. To address this challenge, we employ preference alignment to discern complex relationships between structure and interaction landscapes through comparative analysis of generated samples. Comprehensive experiments demonstrate that ProteinWeaver: (1) generates high-quality, novel protein backbones through versatile domain assembly; (2) outperforms RFdiffusion, the current state-of-the-art in backbone design, by 13\% and 39\% for long-chain proteins; (3) shows the potential for cooperative function design through illustrative case studies. To sum up, by introducing a `divide-and-assembly' paradigm, ProteinWeaver advances protein engineering and opens new avenues for functional protein design.