Using An Artificial Intelligence Algorithm, Researchers at MIT and McMaster University have identified a new Antibiotic that can Kill a Type of Bacteria that is Responsible for Many Drug-Resistant Infections

MIT and McMaster University researchers have utilized artificial intelligence (AI) to discover a new antibiotic that effectively kills drug-resistant bacteria, particularly Acinetobacter baumannii, a species commonly found in hospitals. This bacterium is associated with severe infections such as pneumonia and meningitis, and it is a leading cause of infections among wounded soldiers. The rise of antibiotic-resistant bacteria necessitates the development of new antibiotics, and the use of AI in drug discovery holds great promise.

The researchers employed a machine-learning algorithm to evaluate nearly 7,000 chemical compounds and identify a potential drug that inhibits the growth of Acinetobacter baumannii. The AI algorithm was trained to recognize patterns in extensive data sets and predict the inhibitory properties of chemical compounds. This approach enables the identification of novel antibiotics with distinct chemical structures compared to existing drugs.

In their initial study, the team successfully trained the AI algorithm to identify compounds that could inhibit the growth of E. coli, yielding a molecule named halicin. Halicin demonstrated the ability to kill multiple bacterial species resistant to conventional treatment. Building on this success, the researchers focused on combatting A. baumannii, a significant threat due to its multidrug resistance.

To train their computational model, the researchers exposed A. baumannii to various chemical compounds and observed their inhibitory effects. The AI algorithm analyzed the chemical structures of these compounds and learned to associate specific features with growth inhibition. Next, the algorithm analyzed over 6,000 compounds from the Drug Repurposing Hub at the Broad Institute, quickly identifying a few hundred top candidates. From there, the team selected 240 compounds for experimental testing in the laboratory, prioritizing those with structurally distinct properties from existing antibiotics.

The tests yielded nine potential antibiotics, including one particularly potent compound. Originally investigated as a diabetes drug, this compound effectively kills A. baumannii while leaving other bacterial species unaffected. This narrow spectrum of activity minimizes the risk of bacterial resistance and reduces harm to beneficial gut bacteria that aid in preventing opportunistic infections.

The researchers named the potent antibiotic abaucin and demonstrated its efficacy in treating A. baumannii wound infections in mice. Lab tests confirmed its effectiveness against various drug-resistant strains of A. baumannii isolated from human patients. Further investigations revealed that abaucin interferes with lipoprotein trafficking, a cellular process involved in protein transportation. Notably, abaucin selectively targets A. baumannii despite this process being present in all Gram-negative bacteria. The researchers suggest that subtle differences in how A. baumannii performs lipoprotein trafficking contribute to the drug’s selectivity.

The team is collaborating with McMaster researchers to optimize abaucin’s medicinal properties for potential use in patients. Additionally, they plan to apply their AI modeling approach to identify potential antibiotics for other drug-resistant infections caused by bacteria such as Staphylococcus aureus and Pseudomonas aeruginosa.

The successful application of AI in identifying a novel antibiotic highlights its potential to accelerate and expand the search for effective treatments against drug-resistant bacteria. This research addresses the urgent need for new antibiotics and demonstrates the power of AI in revolutionizing the field of drug discovery.

Check Out The Paper and Reference MIT Article. Don’t forget to join our 23k+ ML SubReddit, Discord Channel, and Email Newsletter, where we share the latest AI research news, cool AI projects, and more. If you have any questions regarding the above article or if we missed anything, feel free to email us at Asif@marktechpost.com

🚀 Check Out 100’s AI Tools in AI Tools Club