The rise of antibiotic-resistant bacteria necessitates the discovery of novel antimicrobial agents. This study leverages a Wasserstein generative adversarial network with gradient penalty (WGAN-GP) to design novel antimicrobial peptides (AMPs). The generated AMPs were evaluated in silico and in vitro. Seven out of eight synthesized peptides showed antibacterial activity, with GAN-pep 3 and GAN-pep 8 demonstrating broad-spectrum activity against antibiotic-resistant strains like methicillin-resistant Staphylococcus aureus and carbapenem-resistant Pseudomonas aeruginosa. GAN-pep 3 showed particularly low minimum inhibitory concentrations (MICs) against all tested bacteria, highlighting its potential as a promising candidate for further development.

T.-T Lin, L.-Y Yang, C.-Y Lin, C.-T Wang, C.-W Lai, Chi-Fong Ko, Francisco Torrens, Antonio Rescifina, Tzu-Tang Lin, Yang-Hsin Shih, Shu-Hwa Chen, C.-F. Shih, Y.-H Chen