AI-assisted design of next-generation antibiotics against multidrug-resistant Staphylococcus aureus

Abstract

Multidrug-resistant (MDR) Staphylococcus aureus remains a leading cause of life-threatening infections worldwide and is designated a high-priority pathogen by the WHO. The accumulation of resistance mechanisms, such as β-lactam insensitivity, reduced vancomycin susceptibility, and multidrug efflux, has limited effective therapies and sustained high morbidity and mortality. Conventional antibiotic discovery is too slow, costly, and inefficient to keep pace with resistance. Artificial intelligence-driven drug design (AIDD) has emerged to address these limitations through high-precision virtual screening, generative de novo design, and multi-parameter property optimization. This review synthesizes the clinical burden and resistance mechanisms of MDR S. aureus , evaluates AIDD technologies spanning data resource curation, resistance prediction, generative design, and structure-based optimization, and examines the structure-activity relationships (SAR) that guide rational anti-staphylococcal design. By integrating AI methodology with antibacterial pharmacology, it illustrates how AI-driven approaches can accelerate the discovery of novel antibiotics against MDR S. aureus and other priority pathogens.