For the First Time Ever, Artificial Intelligence Helps Discover New Antibiotic

The MIT team says that the antibiotic kills some of the most dangerous and resistant strains of bacteria.
Fabienne Lang
The team used a machine-learning algorithm to discover the drug halicinCollins Lab at MIT

Antibiotics are a proven, and rigid, method for ridding our bodies of harmful bacteria. However, many of these bacteria are becoming, or have already become, resistant to many of the drugs we use to fight them.

Researchers keep looking for ways to create new antibiotics, and now a team from MIT has discovered a way of using artificial intelligence (AI) to create a powerful antibiotic. 


Machine learning and antibiotics

Some of the most dangerous and drug-resistant bacteria in the world can now be killed thanks to a new antibiotic developed by AI

The drug operates differently to regular antibiotics and is the first one to be found thanks to using AI.

During tests, the new drug managed to eradicate two out of the three high-priority pathogens that the World Health Organization (WHO) considers "critical." These are antibiotic-resistant and are called Acinetobacter baumannii and Enterobacteriaceae

"In terms of antibiotic discovery, this is absolutely a first," said Regina Barzilay, a senior researcher on the study and specialist in machine learning at MIT.

James Collins, a bioengineer on the MIT team added "I think this is one of the more powerful antibiotics that has been discovered to date. It has remarkable activity against a broad range of antibiotic-resistant pathogens."

If antibiotics become entirely resistant to drugs, by 2050, 10 million lives worldwide could be at risk each year from infections. 

The MIT team trained a deep-learning algorithm to first identify the types of molecules that kill bacteria. Then they looked at a library of over 6,000 compounds currently under investigation for treating a number of human diseases. 

Most Popular

According to Jonathan Stokes, the first author of this study, it only took a matter of hours for the algorithm to assess the compounds, and produce some potential antibiotics. Stokes said that "Being able to perform these experiments in the computer dramatically reduces the time and cost to look at these compounds."

The team named one of the newly discovered antibiotics "halicin," after Hal from A Space Odyssey. The researchers were able to treat a number of drug-resistant infections thanks to halicin.

The team will now focus on making the algorithm find antibiotics that are more selective in the bacteria they eradicate. Ultimately, the plan is to use this algorithm to design antibiotics from the bottom up.

The team's study was published in the journal Cell on Thursday.

message circleSHOW COMMENT (1)chevron