Deep learning-guided discovery of a narrow-spectrum antibiotic against A. baumannii

Abstract

A deep neural network was trained with a data set containing information on over ~7500 different molecules and their effectiveness in inhibiting growth of Acinetobacter baumannii in vitro. Predictions were performed on the Drug Repurposing Hub, and through this approach, Abaucin was discovered to show narrow-spectrum activity against A. baumannii which mitigates issues of resistance mechanisms. .

Further investigations revealed that Abaucin perturbs lipoprotein trafficking through a mechanism involving LolE. This work highlights the utility of machine learning in antibiotic discovery and describes a promising lead with targeted activity against a challenging Gram-negative pathogen.

Applications

Abaucin represents a novel lead molecule that can address untreatable clinical strains of A. baumannii. Ongoing development includes:

• Increasing the size of the A. baumannii antibiotic training dataset
• Performing predictions on larger in silico chemical libraries
• Developing analogs of Abaucin with enhanced potency in vitro

Benefits

• Addresses need for structurally and functionally novel antibiotics for A. baumannii
• Species-selective antibiotics hold promise to limit the horizontal dissemination of resistance determinants
• The machine learning-guided discovery of abaucin highlights the utility of algorithmic approaches to discover novel antibacterial molecules

References:

Liu, G., Catacutan, D.B., Rathod, K. et al. Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii. Nat Chem Biol (2023). https://doi.org/10.1038/s41589-023-01349-8