PPDiff: Diffusing in Hybrid Sequence-Structure Space for Protein-Protein Complex Design

TL;DR

PPDiff is a novel diffusion-based method that jointly designs protein sequences and structures for high-affinity binders to arbitrary targets, significantly advancing protein-protein complex design without extensive lab testing.

Contribution

The paper introduces PPDiff, a diffusion model with a new SSINC network for joint sequence-structure design, outperforming baselines on large-scale protein complex datasets.

Findings

Achieved success rates of 50.00%, 23.16%, and 16.89% on different tasks.

Curated PPBench dataset with over 700,000 complexes from PDB.

Model outperforms baseline methods in protein-protein complex design.

Abstract

Designing protein-binding proteins with high affinity is critical in biomedical research and biotechnology. Despite recent advancements targeting specific proteins, the ability to create high-affinity binders for arbitrary protein targets on demand, without extensive rounds of wet-lab testing, remains a significant challenge. Here, we introduce PPDiff, a diffusion model to jointly design the sequence and structure of binders for arbitrary protein targets in a non-autoregressive manner. PPDiffbuilds upon our developed Sequence Structure Interleaving Network with Causal attention layers (SSINC), which integrates interleaved self-attention layers to capture global amino acid correlations, k-nearest neighbor (kNN) equivariant graph layers to model local interactions in three-dimensional (3D) space, and causal attention layers to simplify the intricate interdependencies within the protein sequence. To assess PPDiff, we curate PPBench, a general protein-protein complex dataset comprising 706,360 complexes from the Protein Data Bank (PDB). The model is pretrained on PPBenchand finetuned on two real-world applications: target-protein mini-binder complex design and antigen-antibody complex design. PPDiffconsistently surpasses baseline methods, achieving success rates of 50.00%, 23.16%, and 16.89% for the pretraining task and the two downstream applications, respectively. The code, data and models are available at https://github.com/JocelynSong/PPDiff.