SeedProteo: Accurate De Novo All-Atom Design of Protein Binders

TL;DR

SeedProteo is a diffusion-based model for de novo all-atom protein design that achieves state-of-the-art results in binder design and demonstrates high success rates and binding affinities validated by wet-lab experiments.

Contribution

It repurposes a folding architecture into a generative design framework with self-conditioning, advancing protein binder design capabilities.

Findings

Superior length generalization and structural diversity in unconditional generation

Highest in-silico success rates among open-source methods for binder design

Achieved 70-80% hit rates and picomolar binding affinities in wet-lab validation

Abstract

We present SeedProteo, a diffusion-based model for de novo all-atom protein design. We demonstrate how to repurpose a cutting-edge folding architecture into a powerful generative design framework by effectively integrating self-conditioning features. Extensive benchmarks highlight the model's capabilities across two distinct tasks: in unconditional generation, SeedProteo exhibits superior length generalization and structural diversity, maintaining robustness for long sequences and complex topologies; in binder design, it achieves state-of-the-art performance among open-source methods, attaining the highest in-silico design success rates, structural diversity and novelty. Finally, we validate SeedProteo through wet-lab assays on two therapeutic targets, achieving hit rates of 70%-80% and picomolar-level binding affinities, establishing leading results. To facilitate community adoption, we provide public access to SeedProteo via a webserver (https://seedfold.io/proteinDesign).