HomeAI News
Scientists use AI to discover antibiotics that kill 'superbugs'

Scientists use AI to discover antibiotics that kill 'superbugs'

Hayo News
Hayo News
May 29th, 2023
View OriginalTranslated by Google

Scientist Denise Catacutan is working on an experimental antibiotic discovered with the help of artificial intelligence

Using artificial intelligence (AI), scientists in the United States and Canada have discovered that a new antibiotic, Abaucin, can kill the deadly superbug Acinetobacter baumannii . This new antibiotic has yet to be further experimentally verified. It is expected that the first batch of AI antibiotics may not be available for prescription until 2030, but this study has shown the great potential of AI in the fields of science and medicine.

According to the latest research published in the scientific journal Nature Chemical Biology, a group of scientists from McMaster University in Canada and the Massachusetts Institute of Technology in the United States used AI algorithms to screen thousands of antibacterial molecules in an attempt to predict new structural categories, This allowed the researchers to identify a new antibacterial compound, Abaucin, that could be used against Acinetobacter baumannii.

The scientists then tested the new antimicrobial compound against Acinetobacter baumannii in a wound infection model in mice and found that the molecule inhibited the infection.

Stokes, an assistant professor in the Department of Biomedicine and Biochemistry at McMaster University who helped lead the study, said: "This work validates the benefits of machine learning in finding new antibiotics. With AI, we can quickly explore huge chemical field, greatly increasing the chances of discovering new antibacterial molecules.”

In the study, the researchers first trained the AI to learn the chemical properties of drugs that work against Acinetobacter baumannii. They then asked the AI to sift through a list of 6,680 potential compounds, and found the potent antibiotic Abaucin.

The results of the study showed that it only took an hour and a half for the AI to produce a condensed list of hundreds of compounds, and then 240 of them were tested in the laboratory, and finally 9 potential antibiotics were found, and then these 9 antibiotics were tested in mice. Abaucin finally came to the fore when tested in a human wound infection model.

However, improvements and clinical trials of Abaucin are still in progress. It is expected that the first AI antibiotics may not be available for prescription until 2030.

Most notably, this antibiotic is ineffective against other bacteria and only works against Acinetobacter baumannii. This precise mode of action could help reduce the emergence of drug resistance and potentially reduce side effects.

The researchers emphasize that the power of AI can screen tens of millions of potential compounds, which cannot be achieved manually. Therefore, AI has the potential to greatly improve the speed and efficiency with which the medical community can discover new classes of antibiotics.

A. baumannii can survive for extended periods of time in environmental services and shared equipment, and can be transmitted through contaminated hands. In addition to bloodstream infections, Acinetobacter baumannii can also cause urinary tract and lung infections.

The bacteria are inherently able to find new ways to resist treatment and can pass on genetic material that makes other bacteria resistant, the WHO said.


no dataCoffee time! Feel free to comment