MAC-AMP: A Closed-Loop Multi-Agent Collaboration System for Multi-Objective Antimicrobial Peptide Design.

This paper introduces MAC-AMP, a closed-loop multi-agent collaboration (MAC) system for multi-objective antimicrobial peptide (AMP) design, addressing the global health threat of antimicrobial resistance. Existing AI models struggle to balance key goals like activity, toxicity, and novelty. MAC-AMP implements an autonomous simulated peer review-adaptive reinforcement learning framework that designs novel AMPs, outperforming other generative models in antibacterial activity, toxicity reduction, and structural reliability.