Thanks to artificial intelligence, an antibiotic kills "superbugs"

 Scientists have resorted to artificial intelligence to come up with a new antibiotic that can kill deadly types of "superbugs", in a move that, if successful, could constitute a breakthrough in the world of medicine.

Fight germs

• Although antibiotics kill bacteria, there has been a shortage of new drugs for decades, and bacteria have become more difficult to treat as they develop resistance to the available drugs.
• It is estimated that more than a million people die annually from infections that are resistant to antibiotic treatment.
• Researchers at Canada's McMaster University focused on one of the most problematic bacteria, Acinetobacter baumannii, which can infect wounds and cause pneumonia.
• This bacteria is one of 3 "superbug" species that the World Health Organization has said is a "serious threat".
• These bacteria are often resistant to antibiotics and are a problem in hospitals and nursing homes, where they can live on surfaces and medical equipment.

How was artificial intelligence exploited?

• Artificial intelligence has helped narrow the thousands of potential chemicals down to a handful that can be tested in a lab.
• The result was a powerful experimental antibiotic called abaucin, which will need more testing before it can be used.
• Researchers in Canada and the United States say that AI "has the potential to dramatically accelerate the discovery of new drugs".

practical steps

• To find a new antibiotic, the researchers first had to train the AI. They took thousands of drugs, where the exact chemical makeup was known, and manually tested them on Acinetobacter baumannii to see which could slow them down or kill them.
• This information was fed into the AI ​​system, so it could learn the chemical properties of the drugs that could attack these superbugs.
• Then, the artificial intelligence was unleashed to determine the efficacy of substances from a list of 6,680 compounds whose efficacy was unknown.
• The results, published in Nature Chemical Biology, show that the AI ​​"took an hour and a half to produce a shortlist".
• The researchers tested 240 in the lab, and found 9 potential antibiotics, one of which was the incredibly powerful antibiotic
• Laboratory experiments showed that the antibiotic was able to treat infected wounds in mice, and was able to kill samples of baumannii bacteria.

What then?

• The next step is to produce an effective drug in the laboratory, and then to conduct clinical trials. It is expected that the first antibiotics to be developed by artificial intelligence will take until the year 2030 to become available for prescription.
• It was strange that this experimental antibiotic had no effect on other types of bacteria, and only worked on. baumannii.
• Many antibiotics kill bacteria indiscriminately. Researchers believe that abaucin's subtlety will make it more difficult for drug resistance to emerge, and may lead to fewer side effects.

Commenting on this revolutionary step, Professor James Collins, from the Massachusetts Institute of Technology, said: "This result supports the hypothesis that artificial intelligence can significantly accelerate and expand our search for new antibiotics." He added, "I'm excited because this work shows we can use AI to help fight pathogens like baumannii."