“We wanted to develop a platform that would allow us to harness the power of artificial intelligence to usher in a new age of antibiotic drug discovery,” says James Collins, a member of the researchers from MIT. “Our approach revealed this amazing molecule which is arguably one of the more powerful antibiotics that has been discovered.”
The algorithm was trained using 2,500 molecules, which included approximately 1,700 drugs approved by the FDA and 800 natural products with bioactives. This was then tested on Broad Institute’s Drug Repurposing Hub collection of 6,000 compounds, which led to the discovery of the antibiotic.
The molecule ‘halicin’—named after the AI system in the classic sci-fi film, 2001: A Space Odyssey—is now under study as a potential drug to combat diabetes. Further testing was done against isolated bacteria strains and was found to successfully kill most, such as Clostridium difficile, Acinetobacter baumannii, and Mycobacterium tuberculosis. It was only unable to kill Pseudomonas aeruginosa, a lung pathogen that is generally challenging to treat. However, these were only done in petri dishes.
To test how effective the antibiotic could be when used on living animals, further tests were conducted on mice carrying A. baumannii. This is a bacterium that has become prevalent among U.S. soldiers stationed in Iraq and Afghanistan. While typically resistant to existing antibiotics, the application of halicin was able to completely eliminate the infection within 24 hours.
The antibiotics industry has been slow to progress over recent years, with minimal drugs being developed. While the use of predictive computer programs in drug discovery isn’t a new phenomenon, existing models are still limited. Researchers from MIT aimed to design a machine-learning-based model capable of being “trained to analyze the molecular structures of compounds and correlate them with particular traits, such as the ability to kill bacteria.”
“We’re facing a growing crisis around antibiotic resistance, and this situation is being generated by both an increasing number of pathogens becoming resistant to existing antibiotics, and an anemic pipeline in the biotech and pharmaceutical industries for new antibiotics,” says Collins.
The team is currently working on using the program to develop more novel antibiotics. They also added that they will continue training the algorithm to further optimize its targeting capabilities.
For more news and stories, check out the role of AI in medicine here.