菌株特异性抗菌肽:生物学基础、智能设计与挑战OA

Broad-spectrum antimicrobial peptide(AMP),while effective in eliminating pathogenic microorganisms,inevitably disturbs host-associated microbial communities and disrupts microecological homeostasis.This intrinsic limitation restricts their application in precise microbial regulation within complex biological systems.Artificial intelligence(AI)is utilized to learn microbial recognition patterns encoded within the physicochemical properties of AMP sequences.By precisely designing and modulating features such as charge distribution,hydrophobicity,spatial conformation and specific structural motifs,AI-driven approaches may enable the development of strain-specific antimicrobial strategies for targeted microbial regulation.Based on this perspective,the biological foundations underlying the selective activity of AMP were systematically summarized and the emerging applications of AI in the design of strain-specific AMP were further discussed.Finally,the major challenges facing AI-driven strain-specific AMP design were outlined,aiming to provide theoretical insights for advancing antimicrobial peptide research from"broad-spectrum killing"toward"precision regulation".