Deep learning helps fight against antibiotic resistance

Area of Science:

• Medicinal Chemistry
• Computational Biology
• Infectious Diseases

Background:

• Antibiotic resistance is a growing global health threat.
• Novel therapeutic strategies are urgently needed to combat resistant bacterial infections.

Purpose of the Study:

• To develop and apply a generative deep learning platform for identifying novel antibacterial compounds.
• To discover lead compounds targeting multidrug-resistant bacterial pathogens.

Main Methods:

• Utilized a generative deep learning platform combining graph neural network (GNN)-based fragment screening.
• Integrated de novo molecular design for lead compound identification.
• Evaluated compound activity in vivo against resistant bacterial strains.

Main Results:

• Identified two lead compounds, NG1 and DN1.
• Demonstrated potent in vivo activity of NG1 and DN1 against multidrug-resistant Neisseria gonorrhoeae.
• Showcased potent in vivo activity of NG1 and DN1 against methicillin-resistant Staphylococcus aureus.

Conclusions:

• The generative deep learning platform is effective for discovering novel antibacterial agents.
• NG1 and DN1 represent promising candidates for treating infections caused by resistant bacteria.
• This AI-driven approach accelerates antibiotic discovery for critical pathogens.