Mason Cheng

Leveraging Machine Learning and Zinc-Aware Docking to Discover Natural Inhibitors of EptA Resistance Enzymes

Started June 26, 2025

Abstract or project description

Mason is conducting a computational research project focused on identifying natural product–derived small molecules that can inhibit the EptA enzyme, a key phosphoethanolamine transferase responsible for polymyxin resistance in Gram-negative bacteria. His project integrates machine learning and molecular docking techniques to screen thousands of compounds and prioritize those most likely to bind to the active site of EptA, especially its zinc-coordinating region. He has compiled a training dataset of known active and inactive compounds, developed predictive models, and validated them using structure-based docking simulations. Mason’s work specifically targets cross-species inhibition, testing against EptA homologs from clinically important pathogens like Neisseria meningitidis, E. coli, Salmonella enterica, and Klebsiella pneumoniae. This project has real-world relevance, as it contributes to the discovery of novel antibiotic adjuvants that could help restore the efficacy of last-resort drugs against multidrug-resistant bacteria.