SOAPIA: Siamese-Guided Generation of Off Target-Avoiding Protein Interactions with High Target Affinity

Sophia Vincoff, Oscar Davis, Ismail Ilkan Ceylan, Alexander Tong, Joey Bose, Pranam Chatterjee

Keywords: multi-objective optimization, protein design, peptide binders, discrete diffusion, Monte Carlo Tree Search, fusion oncoproteins, Siamese networks, biologics, off-target avoidance, affinity prediction, DPLM, AlphaFold-Multimer, rare diseases, therapeutic peptides

TL;DR: We introduce SOAPIA, a dual-guided diffusion framework that designs peptide binders with high affinity and specificity, including to fusion oncoproteins.

Abstract: Therapeutic molecules must selectively interact with a target protein while avoiding structurally or functionally similar off-targets. However, no existing generative strategy explicitly optimizes both target affinity and off-target avoidance. To address this, we introduce SOAPIA, a framework for the Siamese-guided generation of Off-target-Avoiding Protein Interactions with high target Affinity. SOAPIA generates de novo peptide binders by steering the generative process of a Diffusion Protein Language Model (DPLM) using a multi-objective Monte Carlo Tree Search (MCTS). Affinity is optimized via a pre-trained predictor, while specificity is enforced using a Siamese model trained with an adaptive Log-Sum-Exp Decoy Loss. This dual-guidance scheme enables Pareto-efficient exploration of discrete sequence space without gradient access. In benchmarks across 17 fusion oncoproteins, SOAPIA consistently identifies binders with strong affinity and high selectivity for over 75% of targets. For multiple clinically relevant targets, SOAPIA generated peptides that preferentially bind the fusion by engaging both its head and tail domains, while avoiding the wild-type counterparts. These results underscore SOAPIA’s promise for designing safe, specific biologics for fusion-driven cancers and other rare, currently untreatable diseases.

Submission Number: 41