A I-designed compounds can kill drug-resistant bacteria

With help from artificial intelligence, MIT researchers have designed novel antibiotics that can combat two hard-to-treat bacteria: multi-drug-­resistant Neisseria gonorrhoeae and Staphylococcus aureus (MRSA).

The team used two approaches. First, they directed generative AI to design molecules based on a chemical fragment their model had predicted would show antimicrobial activity, and second, they let the algorithms generate molecules without constraints. They designed more than 36 million possible compounds this way and computationally screened them for antimicrobial properties. 

The top candidates they discovered are structurally distinct from any existing antibiotics, and they appear to work by novel mechanisms that disrupt bacterial cell membranes. This makes them less vulnerable to antibiotic resistance, a growing problem: It is estimated that drug-­resistant bacterial infections cause nearly 5 million deaths per year worldwide.

Now that they can generate and evaluate compounds that have never been seen before, the researchers hope they can use the same strategy to identify and design drugs that attack other species of bacteria.

“We’re excited about the new possibilities that this project opens up for antibiotics development,” says James Collins, a professor of biological engineering and the senior author of the study. “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.”