Thanks to the help ofartificial intelligence (Ai) we found in a very short time, in just two hours, a antibiotic which could be used to fight the superbug Acinetobacter baumanniiidentified by the World Health Organization (WHO) as one of the most antibiotic resistant bacteria dangerous in the world. A research team coordinated by McMaster University and the Massachusetts Institute of Technology succeeded in doing so, according to which the newly used AI algorithm could provide valid help to combat theantibiotic resistance and accelerate the discovery of other antibiotics effective in treating many other multidrug-resistant bacteria. The study was published in Nature Chemical Biology.
The dangers of the superbug
Notoriously difficult to eradicate, A. baumannii is a drug-resistant pathogen, capable of picking up the dna from other species of bacteria, including antibiotic resistance genes. It is mainly found in the hospital environmentswhere it can survive on surfaces for long periods, causing pneumonia, meningitis and serious infections in patients that can lead to death. “Acinetobacter can survive on hospital handles and equipment for long periods of time and can absorb i genes of antibiotic resistance from its environment”explains author Jonathan Stokes.
View more
Unlike current methods to find new antibiotics against superbugs, which are time consuming and expensive, modern approaches by machine learning they can access hundreds of millions of molecules with antibacterial properties, and therefore be useful in identifying new classes of bacteria faster drugs. Specifically in the new study, the researchers identified the new molecule among about 7 thousand potential compounds using a machine learning model that they trained to evaluate which chemical characteristics were associated with the inhibition of the growth of A. baumannii. “This work confirms the benefits of machine learning in the search for new antibiotics”Stokes says. “Using theartificial intelligencewe can search faster and significantly increase the chances of discovering fundamentally new antibacterial molecules.”
The advantages of the new antibiotic
It has shown promise so far both in laboratory tests what up animal modelsthe. It was also particularly promising as targets only A. baumannii. Indeed, as the researchers explain, this characteristic is crucial: it means that the pathogen is less likely to rapidly develop drug resistance and could lead to more precise and effective treatments. In fact, most antibiotics are defined as broad-spectrum, which means that they kill all bacteria, even those of the microbiome intestinal. “We know that the antibiotics broad spectrum are suboptimal and that pathogens have the ability to evolve and adapt to every trick we throw at them”conclude Stokes. “AI offers us the opportunity to dramatically increase the rate at which we discover new antibiotics, and we can do it at a reduced cost.”