Scientists use AI to discover new antibiotic to treat deadly superbug | Artificial Intelligence (AI)

Artificial Intelligence (AI)

AI used to discover abaucin, an effective drug against A baumanniibacteria that can cause dangerous infections

Thu May 25, 2023 3:54 PM EDT

Scientists using artificial intelligence have discovered a new antibiotic that can kill a deadly superbug.

According to a new study published Thursday in the scientific journal Nature Chemical Biology, a group of scientists from McMaster University and the Massachusetts Institute of Technology has discovered a new antibiotic that can be used to kill a deadly hospital bacteria.

The super bug in question is Acinetobacter baumanniiwhich the World Health Organization has classified as a “critical” threat among its “priority pathogens” – a group of bacterial families that pose the “greatest threat” to human health.

According to the WHO, the bacteria have a built-in ability to find new ways to resist treatment and can pass on genetic material that allows other bacteria to become resistant to drugs as well.

A baumannii poses a threat to hospitals, nursing homes and patients requiring ventilators and blood catheters, as well as those who have open wounds from surgery.

The bacteria can live on environmental services and shared equipment for long periods of time and can often be spread through contaminated hands. In addition to blood infections, A baumannii can cause infections in the urinary tract and lungs.

According to the Centers for Disease Control and Prevention, the bacteria can also “colonize,” or live inside a patient without causing infections or symptoms.

Thursday’s study found that researchers used an AI algorithm to screen thousands of antibacterial molecules in an attempt to predict new structural classes. As a result of the AI ​​screening, researchers were able to identify a new antibacterial compound they called abaucin.

“We had a bunch of data that just told us which chemicals could kill some bacteria and which ones could not. My job was to train this model, and all this model was going to do is essentially tell us whether new molecules will have antibacterial properties or not,” said Gary Liu, a graduate student from MacMaster University who is working on the research. worked.

“In fact, as a result, we are able to increase the efficiency of the drug discovery pipeline and … hone in on all the molecules that we really want to care about,” he added.

After scientists trained the AI ​​model, they used it to analyze 6,680 compounds it hadn’t encountered before. The analysis took an hour and a half and ultimately yielded several hundred compounds, 240 of which were subsequently tested in a laboratory. Laboratory tests eventually revealed nine potential antibiotics, including abaucine.

The scientists then tested the new molecule against A baumannii in a mouse wound infection model and found that the molecule suppressed infection.

“This work validates the benefits of machine learning in the search for new antibiotics,” said Jonathan Stokes, an assistant professor in McMaster University’s department of biomedical and biochemistry who helped lead the study.

“Using AI, we can quickly explore large swaths of chemical space, greatly increasing the likelihood of discovering fundamentally new antibacterial molecules,” he said.

“We know that broad-spectrum antibiotics are not optimal and that pathogens have the ability to evolve and adapt to whatever trick we play on them… AI methods give us the opportunity to vastly increase the speed at which we discover new antibiotics. increase, and we can do it at a lower cost. This is an important avenue for research into new antibiotics,” he added.

Leave a comment