Scientists use AI to find promising new antibiotics to fight the evasive

Editor’s attention: Embargoed by the journal Nature Chemical Biology until Thursday, May 25 at 11 Easter

Hamilton, ON, May 25, 2023 – Scientists at McMaster University and the Massachusetts Institute of Technology have used artificial intelligence to discover new antibiotics that could be used to fight deadly, drug-resistant pathogens that are attacking susceptible hospital patients.

The process they used could also accelerate the discovery of other antibiotics to treat many other challenging bacteria.

Researchers are responding to the urgent need for new drugs to treat Acinetobacter baumannii, identified by the World Health Organization as one of the most dangerous antibiotic-resistant bacteria in the world. Known as difficult to eradicate, A. baumannii can cause pneumonia, meningitis and infect wounds, all of which can result in death.

A. baumanni it is usually found in hospitals, where it can survive on surfaces for long periods of time. Pathogens are able to take up DNA from other bacterial species in their environment, including antibiotic-resistant genes.

In the study, published today in the journal Natural Chemical Biology, Researchers report they used an artificial intelligence algorithm to predict a new structural class of antibacterial molecules, and identified a new antibacterial compound, which they named abaucin.

Find new antibiotics to fight A. baumannii through conventional screening has been challenging. Traditional methods are time consuming, expensive, and limited in scope.

Modern algorithmic approaches can access hundreds of millions, perhaps billions of molecules with antibacterial properties.

“This work validates the benefits of machine learning in the search for new antibiotics,” said Jonathan Stokes, lead author of the paper and assistant professor in McMaster’s Department of Biomedical & Biochemistry, who conducted the work with James J. Collins, a professor of medical engineering and science at MIT, and student graduates McMaster, Gary Liu and Denise Catacutan.

“By using AI, we can quickly explore large regions of the chemical space, significantly increasing the chances of discovering fundamentally new antibacterial molecules,” said Stokes, who is a Fellow of McMaster’s Global Nexus School for Pandemic Prevention and Response.

“AI approaches to drug discovery are here to stay and will continue to be refined,” said Collins, chair of the Life Sciences faculty at the MIT Abdul Latif Jameel Clinic for Machine Learning in Health. “We know the algorithmic model works, now it’s just a matter of widely adopting this method to find new, more efficient and less expensive antibiotics.”

Abaucin holds great promise, report the researchers, because it only targets A. baumanniian important finding that means pathogens are less likely to quickly develop drug resistance, and that can lead to more appropriate and effective treatments.

Most antibiotics are broad-spectrum, meaning they kill all the bacteria, disrupting the gut microbiome, which opens the door to a number of serious infections, including Difficult.

“We know broad-spectrum antibiotics are not optimal and that pathogens have the ability to evolve and adapt to every trick we throw at them,” said Stokes. “AI methods give us the opportunity to significantly increase the speed of discovery of new antibiotics, and we can do this at a lower cost. This is an important avenue of exploration for new antibiotic drugs.”

Editor’s note: High resolution photos with credit information may be downloaded for use hereink: https://photos.app.goo.gl/gwGU2YQ8V4YRsPbN8