Scientists Develop New Antibiotic To Kill Superbug Using AI

Machine learning enabled scientists to rapidly screen thousands of molecules to find a new antibacterial compound.

By Mat Di Salvo

2 min read

Artificial intelligence might worry some—but the tech is already doing wonders for humanity. Scientists have used AI to discover a new antibiotic that can fight a drug-resistant superbug deemed a “critical” threat by the World Health Organization.

Using machine learning, scientists from McMaster University and the Massachusetts Institute of Technology were able to find a new antibiotic to kill Acinetobacter baumannii, a bacteria often found in hospitals that can cause serious infections, according to a study published last week.

Acinetobacter baumannii can spread short distances through the air in water droplets but is most commonly transmitted person-to-person, via the hands of healthcare workers.

The study's lead author Jonathan Stokes. Image: McMaster University

It has become resistant to most antibiotics and the World Health Organization said back in 2017 that new antibiotics were “urgently needed” to fight it. 

But the study said that AI may be well on the way to achieving that goal. 

“Machine learning approaches allow for the rapid exploration of chemical space, increasing the probability of discovering new chemical matter with antibacterial activity,” the study published in the journal Nature Chemical Biology

Machine learning is a branch of AI where computers learn automatically from past data. When fighting disease, the technology can be used to screen hundreds of millions to billions of molecules with the suitable antibacterial properties. 

In this case, the scientists screened 7,500 molecules to find a new antibacterial compound named abaucin. It will need further testing before it is used but laboratory tests showed that it can treat infected wounds on mice. 

They added that it was a “promising lead” and “highlights the utility of machine learning in antibiotic discovery.”

Get crypto news straight to your inbox--

sign up for the Decrypt Daily below. (It’s free).

Recommended News