AI discovers new antibiotic to kill deadly superbug
Scientists have used artificial intelligence (AI) to discover a new antibiotic that can kill one of the most dangerous superbugs in the world.
The superbug is called Acinetobacter baumannii and it can cause infections in wounds and lungs. It is often resistant to multiple antibiotics and can spread easily in hospitals and care homes. The World Health Organization has classified it as a "critical" threat among its "priority pathogens" – a group of bacteria families that pose the greatest risk to human health.
The researchers from McMaster University in Canada and the Massachusetts Institute of Technology in the US used an AI algorithm to screen thousands of antibacterial molecules and predict new ones. The AI helped them narrow down the search from 6,680 compounds to nine potential antibiotics in just an hour and half.
One of them was a potent, experimental antibiotic called abaucin, which showed promising results in laboratory tests. It was able to treat infected wounds in mice and kill A. baumannii samples from patients. The researchers say abaucin is a new structural class of antibiotic that works only on A. baumannii and does not affect other bacteria.
However, abaucin is not ready to be used yet. It will need further tests and clinical trials before it can be prescribed to humans. The researchers estimate that the first AI antibiotics could take until 2030 to become available.
The researchers say that AI has the power to massively accelerate the discovery of new drugs and overcome the challenge of antibiotic resistance. They hope that their method can be applied to other superbugs and diseases in the future.
The study was published on Thursday in the science journal Nature Chemical Biology.
Scientists have used artificial intelligence (AI) to discover a new antibiotic that can kill one of the most dangerous superbugs in the world.
The superbug is called Acinetobacter baumannii and it can cause infections in wounds and lungs. It is often resistant to multiple antibiotics and can spread easily in hospitals and care homes. The World Health Organization has classified it as a "critical" threat among its "priority pathogens" – a group of bacteria families that pose the greatest risk to human health.
The researchers from McMaster University in Canada and the Massachusetts Institute of Technology in the US used an AI algorithm to screen thousands of antibacterial molecules and predict new ones. The AI helped them narrow down the search from 6,680 compounds to nine potential antibiotics in just an hour and half.
One of them was a potent, experimental antibiotic called abaucin, which showed promising results in laboratory tests. It was able to treat infected wounds in mice and kill A. baumannii samples from patients. The researchers say abaucin is a new structural class of antibiotic that works only on A. baumannii and does not affect other bacteria.
However, abaucin is not ready to be used yet. It will need further tests and clinical trials before it can be prescribed to humans. The researchers estimate that the first AI antibiotics could take until 2030 to become available.
The researchers say that AI has the power to massively accelerate the discovery of new drugs and overcome the challenge of antibiotic resistance. They hope that their method can be applied to other superbugs and diseases in the future.
The study was published on Thursday in the science journal Nature Chemical Biology.
Nature
Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii
Nature Chemical Biology - Using a neural network trained on bacterial growth inhibition data, in silico prediction of molecules with activity against Acinetobacter baumannii led to the...