Antimycobacterial Peptides: From Natural Product Discovery to AI Guided Design

Mycobacterial pathogens remain major global health threats, exacerbated by both rapid acquisition of antibiotic resistance and the formidable drug diffusion barrier presented by the rigid mycomembrane. These challenges have renewed interest in antimycobacterial peptides (AMyPs), a diverse class of short amphiphilic sequences capable of rapidly killing both drug-sensitive and drug-resistant mycobacteria. Beyond their intrinsic potency, AMyPs can synergize with existing antibiotics and exhibit markedly slower resistance development relative to conventional small molecules. In this review, we synthesize recent advances spanning natural bioprospecting, mechanism-guided rational design, and chemical optimization strategies that have yielded increasingly potent and selective AMyP candidates. We further highlight the rapid emergence of artificial intelligence-driven discovery platforms, which leverage machine-learning models trained on curated, mycobacteria-specific data sets to predict and refine novel AMyPs with growing accuracy. Together, these technologic, biologic, and computational advances outline a rapidly expanding landscape for AMyP-based therapeutic development and establish a foundation for next-generation antimycobacterial drug design.