Scientists have used artificial intelligence to discover a new antibiotic capable of killing a deadly superbug, according to a study published in Nature Chemical Biology. Researchers from McMaster University and the Massachusetts Institute of Technology identified the compound, named abaucin, which targets Acinetobacter baumannii, a bacterium classified as a 'critical' threat by the World Health Organization.
Acinetobacter baumannii poses significant risks in hospitals and nursing homes, particularly for patients on ventilators, with blood catheters, or recovering from surgery. The bacteria can survive on surfaces and equipment for extended periods and spread via contaminated hands, causing infections in the blood, urinary tract, and lungs. The WHO lists it among priority pathogens that pose the greatest threat to human health due to its ability to resist treatment and transfer drug-resistance genes.
The research team employed an AI algorithm to screen thousands of antibacterial molecules, predicting new structural classes. Gary Liu, a graduate student at McMaster University, explained that the model was trained to identify which chemicals kill bacteria, enabling efficient drug discovery. The AI analysed 6,680 compounds in 90 minutes, yielding several hundred candidates. Laboratory tests on 240 of these identified nine potential antibiotics, including abaucin.
Further testing in a mouse wound infection model showed that abaucin suppressed the infection. Jonathan Stokes, assistant professor at McMaster University, stated: 'This work validates the benefits of machine learning in the search for new antibiotics. Using AI, we can rapidly explore vast regions of chemical space, significantly increasing the chances of discovering fundamentally new antibacterial molecules.' He added that AI methods reduce costs and accelerate the discovery of new antibiotics, which is crucial as pathogens evolve resistance to existing drugs.
