Surface-based Molecular Design with Multi-modal Flow Matching
This work addresses the challenge of designing therapeutic peptides for undruggable targets by integrating surface information, offering a novel approach that could enhance drug discovery.
The paper tackled the problem of peptide design by incorporating molecular surfaces, which are critical for protein-protein interactions but previously underexplored, and introduced SurfFlow, a surface-based generative algorithm that outperformed full-atom baselines on the PepMerge benchmark.
Therapeutic peptides show promise in targeting previously undruggable binding sites, with recent advancements in deep generative models enabling full-atom peptide co-design for specific protein receptors. However, the critical role of molecular surfaces in protein-protein interactions (PPIs) has been underexplored. To bridge this gap, we propose an omni-design peptides generation paradigm, called SurfFlow, a novel surface-based generative algorithm that enables comprehensive co-design of sequence, structure, and surface for peptides. SurfFlow employs a multi-modality conditional flow matching (CFM) architecture to learn distributions of surface geometries and biochemical properties, enhancing peptide binding accuracy. Evaluated on the comprehensive PepMerge benchmark, SurfFlow consistently outperforms full-atom baselines across all metrics. These results highlight the advantages of considering molecular surfaces in de novo peptide discovery and demonstrate the potential of integrating multiple protein modalities for more effective therapeutic peptide discovery.