A team of US and Canadian scientists have used AI (artificial intelligence) to discover a new superbug-killing antibiotic. This drug-resistant pathogen, Acinetobacter baumannii (A. baumannii) is a serious threat today to vulnerable hospital patients.

The scientists used a process that could also accelerate the discovery of other antibiotics to treat a wide range of nasty bacteria.

The World Health Organization (WHO) has identified A. baumannii as one of the most dangerous antibiotic-resistant bacteria in the world today. The bacterium, which is extremely difficult to eradicate, infects wounds, and causes meningitis and pneumonia. Most infections with this superbug can lead to death.

About A. baumannii

A. baumannii, which is named after bacteriologist Paul Baumann who discovered it in a soil sample in 1968, is found almost exclusively in hospital environments. Even though it has been found in water samples and environmental soil, scientists say they still do not know what its natural environment is.

Image of Acinetobacter baumannii - cluster of bacteria

Wikimedia Commons image adapted by MedicalVocab.com.

It can survive on surfaces for a long time. It can pick up DNA from other types of bacteria in its environment, including drug-resistant genes.

AI faster than traditional methods

The scientists published their findings in the academic journal Nature Chemical Biology (citation below). They have named the new compound that their AI found Abaucin.

Discovering new antibiotics against superbugs like A. baumannii using traditional methods is very costly, time-consuming, and limited in scope. With modern algorithms, it is possible to access hundreds of millions and maybe even billions of molecules with antibacterial qualities.

Lead author, Jonathan Stokes, assistant professor at McMaster’s Department of Biomedicine & Biochemistry, said:

“This work validates the benefits of machine learning in the search for new antibiotics.”

Abaucin’s narrow focus

Abaucin only targets A. baumannii, which could lead to much more precise and effective treatments. The bacterium is also less likely to develop drug resistance.

Most antibiotics act like a scatter gun, wiping out all bacteria, not just the bad ones. This can upset our gut’s natural balance of bacteria, potentially leading to serious infections, such as C. difficile.

Prof. Stokes said:

“We know broad-spectrum antibiotics are suboptimal and that pathogens have the ability to evolve and adjust to every trick we throw at them. AI methods afford us the opportunity to vastly increase the rate at which we discover new antibiotics, and we can do it at a reduced cost. This is an important avenue of exploration for new antibiotic drugs.”

Liu, G., Catacutan, D.B., Rathod, K. et al. (2023). Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii. Nat Chem Biol. (2023)