AI-Assisted Drug Discovery Against Multidrug-Resistant Bacteria

The increasing occurrence of multidrug-resistant (MDR) bacteria, commonly known as superbugs. It is a leading global health threat. The antibiotic discovery pipeline is effectively stagnant due to excessive costs, a long lead time for drug development, and decreased profits for pharmaceutical companies. Artificial intelligence (AI) and machine learning (ML) have proven to be thriving zeitgeists for advancing antimicrobial research through the rapid evaluations of large biological and chemical datasets, predicting antimicrobial activity, identifying novel drug targets, and optimizing pharmacokinetics. This review outlines the various applications of AI-based endeavours in solving the issue of MDR pathogens. These include target identification, virtual screenings, de novo drug design, drug repurposing, optimizing pharmacokinetics, and integrating with experimental systems biology. We will discuss significant discoveries such as halicin and abaucin, as well as limitations including data availability and interpretability. We will explore regulatory aspects and ethical aspects of AI and ML applications, and we will propose future directions for integrating AI and ML in clinical microbiology and personalized medicine to subsume the global antimicrobial resistance (AMR) crisis.