Scientists have used artificial intelligence (AI) to create new potential drugs against what are called superbacteria, stubborn bacterial infections that escape existing treatments.
The AI has reshaped the discovery of drugs in recent years, helping researchers and drug manufacturers to identify promising treatments by accelerating the laborious process of searching for effective compounds likely to be transformed into drugs.
But the scientists of the Massachusetts Institute of Technology (MIT) went further, using AI to generate hypothetical chemical molecules which have not been discovered or that do not yet exist.
The objective was to find completely new ways to fight against antimicrobial resistance, that is to say when bacteria, viruses, fungi or parasites evolve to the point that the drugs designed to kill them are no longer effective, which makes infections more difficult to treat.
The MIT team has targeted the drug -resistant gonorrhea, which the American health authorities call “urgent threat to public health”, and the multidistant golden staphylococcus (MDRSA). The latter includes the gold -resistant golden staphylococcus (STM), which can be contracted in contact with infected persons or contaminated medical equipment.
“We wanted to get rid of everything that could look like an existing antibiotic, in order to help resolve the antimicrobial resistance (RAM) crisis in a fundamentally different way,” said Aarti Krishnan, MIT researcher and one of the study authors.
“By adventing ourselves in unexplored fields of chemical space, our goal was to discover new mechanisms of action,” she added.
Related
The team used AI generative algorithms to create more than 36 million potential compounds and find the best candidates to kill the bacteria.
They identified a fragment which seemed to work well against the Gonorrhea bacteria and, after some additional adjustments, they transformed two of these digital candidates into real compounds.
One of them, which they baptized NG1, turned out to be very effective in killing Gonorrhea bacteria in a laboratory box and in a mouse model.
At the end of a similar process aimed at finding potential treatments for MDRSA, six molecules have proven to be effective against laboratory cultivated bacteria.
The researchers said that these results, published in the journal Cell, could help them create and assess new potential compounds to target other species of bacteria.
Globally, drug -resistant bacterial infections contributed to around 4.71 million deaths in 2021, and this figure is expected to increase in the coming decades.
“Our work shows the power of AI from the point of view of the design of drugs and allows us to exploit much larger chemical spaces which were previously inaccessible,” said James Collins, professor at the MIT and one of the authors of the study, in a press release.
Scientists are now working with organic lighthouse, a non-profit biotechnology company, to continue to test laboratory compounds. If they remain promising, these candidate drugs could possibly be the subject of clinical trials.
“We are excited by the new possibilities that this project opens for the development of antibiotics,” said Professor Collins.