The development of potent and broad-spectrum antimicrobial peptides (AMPs) is crucial to combatting antimicrobial resistance. This study introduces deepAMP, a peptide language-based deep generative framework, to identify such AMPs. Through two rounds of design and cross-optimization-validation, 18 T1-AMPs and 11 T2-AMPs were identified, synthesized, and tested. Over 90% exhibited superior antibacterial activity against Gram-positive and Gram-negative bacteria, with T2-9 showing activity comparable to FDA-approved antibiotics. Selected AMPs significantly reduced resistance in S. aureus and proved effective against P. aeruginosa wound infection in a mouse model. deepAMP accelerates the discovery of effective, broad-spectrum AMPs against drug-resistant bacteria.
