Harnessing AI-powered computational drug design to combat antimicrobial resistance: A comprehensive review of challenges, opportunities, and future directions
P. Santhiya, M P Goutham Bharathi, Rohit R Menon, K. Rajalakshmi, J. Manikandan · In Silico Research in Biomedicine · 2025
Drug design and discovery is a computationally advanced, multidisciplinary process that faces numerous challenges from target identification to clinical trials and final commercialization. These challenges are especially significant in combating antimicrobial resistance (AMR), a global health crisis that undermines the efficacy of current antibiotics. AMR is driven by factors such as overuse and misuse of antibiotics, inadequate infection control, and insufficient investment in novel antimicrobial research. The rapid transmission of resistant pathogens not only thwart clinical intervention but also causes vast economic and social consequences. Traditional drug discovery approaches are often time intensive and expensive. As a result, integrating artificial intelligence (AI) and machine learning (ML) technologies has become essential. Deep learning models such as AlphaFold and ESMFold have revolutionized structural biology by accurately predicting protein structures, thereby accelerating structure-based drug design. AI enables early-stage, high-throughput screening of antimicrobial candidates, models drug-target interactions with high fidelity, and supports the design of narrow-spectrum antibiotics that minimize disruption to the human microbiome. In the AMR context, AI is beginning to deliver on its potential to uncover novel antibiotic classes, predicting bacterial resistance mechanisms, and engineering of more precise therapeutic agents. This article outlines the transformative role of AI in reshaping the drug discovery pipeline, particularly, in addressing the urgent need for effective antimicrobials with particular focus on the fight against AMR, and also explores the challenge of finding new therapeutics as well as finding new pharmacological targets.