PepEVOLVE: Position-Aware Dynamic Peptide Optimization via Group-Relative Advantage

TL;DR

PepEVOLVE is a novel, position-aware framework for dynamic peptide optimization that learns where and how to edit peptides for multiple objectives, outperforming prior static methods in efficiency and quality.

Contribution

It introduces a dynamic, position-aware approach with a multi-armed bandit router and evolving optimization, enhancing peptide lead optimization without pre-specified mutable sites.

Findings

Outperformed PepINVENT in mean score (0.8 vs. 0.6)

Achieved best candidate score of 0.95 vs. 0.87

Converged faster in optimizing permeability and lipophilicity

Abstract

Macrocyclic peptides are an emerging modality that combines biologics-like affinity with small-molecule-like developability, but their vast combinatorial space and multi-parameter objectives make lead optimization slow and challenging. Prior generative approaches such as PepINVENT require chemists to pre-specify mutable positions for optimization, choices that are not always known a priori, and rely on static pretraining and optimization algorithms that limit the model's ability to generalize and effectively optimize peptide sequences. We introduce PepEVOLVE, a position-aware, dynamic framework that learns both where to edit and how to dynamically optimize peptides for multi-objective improvement. PepEVOLVE (i) augments pretraining with dynamic masking and CHUCKLES shifting to improve generalization, (ii) uses a context-free multi-armed bandit router that discovers high-reward residues, and (iii) couples a novel evolving optimization algorithm with group-relative advantage to stabilize reinforcement updates. During in silico evaluations, the router policy reliably learns and concentrates probability on chemically meaningful sites that influence the peptide's properties. On a therapeutically motivated Rev-binding macrocycle benchmark, PepEVOLVE outperformed PepINVENT by reaching higher mean scores (approximately 0.8 vs. 0.6), achieving best candidates with a score of 0.95 (vs. 0.87), and converging in fewer steps under the task of optimizing permeability and lipophilicity with structural constraints. Overall, PepEVOLVE offers a practical, reproducible path to peptide lead optimization when optimal edit sites are unknown, enabling more efficient exploration and improving design quality across multiple objectives.