When an antibiotic fails: MIT scientists are using AI to target “sleeper” bacteria
- Nyquiste
- Apr 8, 2024
- 2 min read
Updated: Feb 19

AI-Powered Discovery of Antibiotics for Dormant Bacteria
Modern antibiotic discovery has faced stagnation since the 1970s, contributing to the growing crisis of antimicrobial resistance. The World Health Organization has identified this as one of the top 10 global public health threats.
Recurring infections often result from bacteria entering a metabolically inactive state, allowing them to evade traditional antibiotics that target active bacteria. These dormant bacteria, or "sleeper" cells, can later reactivate and cause persistent infections.
MIT’s AI Approach to Antibiotic Discovery
A research team at MIT’s Jameel Clinic, led by James J. Collins, has leveraged AI to accelerate antibiotic discovery. In a study published in Cell Chemical Biology, the team demonstrated how machine learning models can efficiently screen compounds lethal to dormant bacteria, significantly reducing the time required for drug discovery.
Breakthrough with Semapimod
Using AI-driven high-throughput screening, researchers identified semapimod, an anti-inflammatory drug previously used for Crohn’s disease, as an effective antibiotic against stationary-phase Escherichia coli and Acinetobacter baumannii over a single weekend.
Key findings:
Membrane Disruption: Semapimod effectively disrupts the outer membrane of Gram-negative bacteria, making them susceptible to antibiotics usually effective only against Gram-positive bacteria.
Overcoming Resistance: Gram-negative bacteria, such as E. coli, A. baumannii, Salmonella, and Pseudomonas, possess highly resistant outer membranes, making new antibiotic discoveries crucial.
Future Implications
By combining AI with molecular biology, researchers can accelerate the identification of novel antibiotics to combat resistant bacterial strains. As Valeri, the study’s lead author, explains, this approach provides a promising strategy for overcoming the challenges of antibiotic resistance, a pressing global health concern.
Conclusion
The MIT team’s innovative AI-driven method represents a significant step forward in tackling antimicrobial resistance. With continued research and technological advancements, AI has the potential to revolutionize antibiotic discovery, offering hope for more effective treatments against persistent infections.
Comments