AI-designed antibiotics clear drug-resistant infections in mice

Researchers in the US used artificial intelligence to discover two new antibiotics that can clear superbugs in mice infected with MRSA and drug-resistant gonorrhoea. 

Most of our current antibiotics were discovered in the mid-20th century, giving bacteria decades to evolve defences against these drugs. Antibiotic-resistant bacterial infections are estimated to cause nearly five million deaths worldwide per year.  

Among the most common resistant bacteria is MRSA (multi-drug-resistant Staphylococcus aureus), dubbed a 'superbug' because it resists multiple types of antibiotics, making it increasingly difficult to treat. 

Researchers at the Massachusetts Institute of Technology, Boston, used AI to design more than 36 million new molecules. Each of these molecules was then analysed by AI, which had been trained to predict antibacterial activity.  

From 24 promising candidates, seven were proven to successfully eliminate MRSA or drug-resistant gonorrhoea (N. gonorrhoeae) in a lab dish. When tested in mice, the two most promising candidates were able to completely clear MRSA skin infections and eliminate most bacteria in drug-resistant gonorrhoea vaginal infections. These new drugs acted by interfering with the cell membranes of bacteria. 

James Collins, senior author of the study published in Cell and professor at MIT, said: “We’re excited about the new possibilities that this project opens up for antibiotics development. Our work shows the power of AI from a drug design standpoint and enables us to exploit much larger chemical spaces that were previously inaccessible.” 

The researchers plan to apply this AI technology to other antibiotic-resistant bacteria that cause deadly diseases, such as tuberculosis.