MIT Scientists discover a Powerful New Antibiotic using Artificial Intelligence

Feb 20, 2020 by Ameer Helles

MIT researchers have discovered a powerful new antibiotic compound using a machine-learning algorithm. The drug killed most of the world's most problematic disease-causing bacteria in laboratory trials, including some strains resistant to all identified antibiotics, which were threatening human existence.

With the ever-growing concern of antibiotic resistance, the discovery and development of new antibiotics are becoming more urgent. Traditional antibiotic discovery approaches can be very expensive and time-consuming, but this new technique could reduce the period from years down to weeks. The computer model, which screens more than a hundred million compounds in a matter of days, is designed to pick potential antibiotics that use different mechanisms to eradicate bacteria than those of existing drugs.

Image source: The Collins Lab at MIT

"We wanted to reveal a platform that allows us to exploit the power of AI to guide in a new age of antibiotic discovery," says James Collins, the Termeer Professor of Medical Engineering and Science in MIT's Institute for Medical Engineering and Science (IMES) and Department of Biological Engineering. "Our approach revealed this amazing molecule, which is arguably one of the major powerful antibiotics that has been discovered."

The scientists behind the study have emphasized that this is only the beginning in terms of this technology. Now that it has been recognized just how valuable a tool this is. The AI aspect can be enhanced continuously so that larger libraries can be screened. This could result in molecules with antimicrobial activity being discovered continually.

The team hopes that this will contribute to a formulation of drugs administered intravenously to a precise location of infection in the future, so as not to interact harmlessly with the bacteria that reside within us.
 

---