AI and New Antibiotics


In medicine, AI has improved everything from surgery to medical research. Now, it may be delivering its best results yet: the development of new antibiotics.

Over time, bacteria have become resistant to treatment. The ones that can overcome multiple antibiotics are called "super bugs". Doctors have gotten better about not over-prescribing antibiotics. But that's less profitable for drug makers, which discourages them from developing new ones. That's why AI could be the answer.

Scientists used machine learning and AI to find new antibiotics against the deadly Acinetobacter baumannii. It's a major problem in hospitals and the military where it causes serious illnesses such as meningitis and pneumonia. Based on the bacterium's structural features, researchers gave the AI nearly seven thousand compounds to analyze. In just a few hours, the learning model picked out just over two hundred that could be new antibiotics. This hugely cut down time in the lab so that they only had to test two hundred forty compounds. Of those, nine could work and one called, abaucin, was a home run.

It was able to target and kill A. baumanni by interfering with lipoprotein trafficking which damages the membranes of the bacterium. Other experiments showed that abaucin prevented the bacterium from establishing wound infections in mice. This work shows us that AI can help us find not only antibiotics but other effective drugs. It could have huge payoffs for our health.

You can now hear additional episodes on many of your favorite podcast providers - visit Buzzsprout to subscribe.

More Information

AI Could Quickly Screen Thousands of Antibiotics to Tackle Superbugs
As the threat of antibiotic resistance increases, new antibiotics are imperative-and AI could widen the pipeline...

Using AI, scientists find a drug that could combat drug-resistant infections
The machine-learning algorithm identified a compound that kills Acinetobacter baumannii, a bacterium that lurks in many hospital settings...

Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii
Acinetobacter baumannii is a nosocomial Gram-negative pathogen that often displays multidrug resistance. Discovering new antibiotics against A. baumannii has proven challenging through conventional screening approaches. Fortunately, machine learning methods allow for the rapid exploration of chemical space, increasing the probability of discovering new antibacterial molecules...