Researchers discover new class of antibiotics against MRSA


Multiresistant pathogens are a major problem. A new class of AI-discovered antibacterials could help.

MIT researchers have used artificial intelligence to discover a new class of compounds capable of killing methicillin-resistant Staphylococcus aureus (MRSA), a drug-resistant bacterium that causes more than 10,000 deaths a year in the US. These compounds have low toxicity to human cells, making them potential drug candidates.

An important aspect of their approach was to use the Monte Carlo tree search algorithm to better explain the deep learning models. This allowed the researchers to understand how the models arrived at their predictions and to identify which substructures of the molecules were likely to be responsible for their antimicrobial activity.

New antibiotic candidates expected to be safe for human cells

To refine the selection of potential drugs, they trained three additional models to predict the toxicity of the compounds to three types of human cells. By combining this data with the predictions of antimicrobial activity, they were able to find compounds that kill microbes while having minimal negative effects on humans.



Specifically, the team screened around 12 million commercially available compounds and identified five different classes with likely activity against MRSA. Two compounds in particular emerged as promising antibiotic candidates, reducing the MRSA population by a factor of 10 in two mouse models. These compounds appear to work by disrupting the bacteria’s ability to maintain an electrochemical gradient across their cell membranes, which is essential for key cellular functions.

Team shares results with nonprofit organization

The research has been shared with Phare Bio, a non-profit organization founded by some researchers involved, which will carry out more detailed analyses of the compounds’ chemical properties and potential clinical applications. Meanwhile, the laboratory is working to develop further drug candidates based on the new study results.

The researchers believe that their AI-based approach can be easily transferred to the discovery of new classes of antibiotics against various pathogens. “We are already leveraging similar approaches based on chemical substructures to design compounds de novo, and of course, we can readily adopt this approach out of the box to discover new classes of antibiotics against different pathogens,” says Felix Wong, a postdoctoral fellow at the Institute for Medical Engineering and Science (IMES) at MIT and Harvard University.

In addition to researchers from MIT and Harvard University, the study also involved researchers from the Leibniz Institute for Polymer Research and the Max Bergmann Center of Biomaterials in Germany.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top