Discovery of a Novel <scp>FabI</scp> Inhibitor as Antibacterial Agents Based on Machine Learning

ABSTRACT Staphylococcus aureus , a pathogen well known for its antibiotic resistance, is recognized by public health agencies as a high‐priority threat due to its remarkable ability to acquire multidrug resistance. The type II fatty acid biosynthesis (FAS‐II) pathway has emerged as a tractable and promising target for antibacterial drug discovery. Within this pathway, inhibition of enoyl‐acyl carrier protein reductase (FabI) can both suppress bacterial growth and restore the susceptibility of multidrug‐resistant strains to existing antibiotics. Although triclosan and isoniazid have been approved for bacterial infections, the variability of clinical outcomes underscores the urgent need to identify novel FabI inhibitors as backup options. Accordingly, we designed a multi‐stage computational strategy to identify FabI inhibitors from our in‐house chemical library, incorporating machine learning (ML)‐based classification models, molecular docking, ADMET property assessment, and clustering analysis. As a result, we characterized a novel chemotype, compound Y4 [N‐(2‐(4‐butoxybenzoyl)hydrazine‐1‐carbonothioyl)propionamide], as a FabI inhibitor with an IC 50 of 9.83 μM. Compound Y4 also demonstrated potent antibacterial activity against Staphylococcus aureus (MIC = 2 μg/mL) in broth microdilution assays, and molecular dynamics simulations were conducted to characterize its interactions with FabI and inform future structural optimization.