Using AI, MIT researchers identify a new class of antibiotic candidates

Dec 22, 2023 ·

MIT researchers have developed a new AI system that could help lead to the discovery of potential antibiotics, helping scientists find new treatments for infections. The AI system works by analyzing the structure of molecules to identify those that could disrupt the growth and replication of bacteria.

By analyzing thousands of molecules, the system can quickly identify which ones might have antibiotic properties, based on the way the molecules interact with bacterial proteins. This process is much faster than traditional laboratory testing, which requires a lengthy process to analyze only a few molecules at a time.

The team tested their system on a database of compounds that are currently in the early stages of drug development and found that it was able to successfully identify potential antibiotics at a rate of 83%. The system was also able to distinguish between molecules that had different levels of antibacterial activity, allowing it to suggest more effective candidates.

This technology could potentially lead to the discovery of novel antibiotics at a much faster pace. It could also reduce the amount of time and resources needed to bring a new antibiotic to market. Furthermore, by focusing on a larger pool of molecules, the system could uncover previously unknown antibiotic molecules that have yet to be discovered.

In conclusion, the MIT research team's new AI system may provide an efficient and effective way to identify potential antibiotic candidates. By quickly analyzing thousands of molecules, the system could find novel antibiotics, increasing chances of success in finding treatments for bacterial infections.