Scientists from McMaster University and Massachusetts Institute of Technology used AI to discover a new antibiotic capable of combating a deadly superbug, abaucin.

To identify a new antibiotic, researchers trained the AI by testing thousands of drugs with known chemical structures on Acinetobacter baumannii to identify those that could slow down or eliminate the bacteria. This information was used to train the AI to recognize the chemical features of drugs effective against the problematic bacterium. The AI was then deployed to evaluate a list of 6,680 compounds with unknown efficacy, producing a shortlist in just an hour and a half.

After laboratory experiments on 240 compounds, nine potential antibiotics were discovered, including the highly potent abaucin. This experimental antibiotic demonstrated the ability to treat infected wounds in mice and kill samples of A. baumannii taken from patients. However, the researchers emphasize that further work is needed to refine the drug in the lab and conduct clinical trials.

The unique aspect of abaucin is its precision, as it only affects A. baumannii and has no impact on other species of bacteria. This specificity could make it more challenging for drug resistance to emerge and potentially reduce side effects.

Article written by James Gallagher