UMass team uses AI, PAC-MAN to speed TB drug discovery
This digest was compiled by AI from multiple sources — links to the originals are below.
Researchers at the University of Massachusetts Amherst have developed two techniques that combine AI and a molecular tool called PAC-MAN to accelerate the discovery of drugs against tuberculosis, which killed 1.23 million people in 2024. The methods target the bacterium's impenetrable outer membrane, a key barrier to existing antibiotics. The findings were published in a peer-reviewed journal.
The Challenge
Tuberculosis, caused by Mycobacterium tuberculosis (Mtb), is the world's deadliest single-agent infection, with 1.23 million deaths in 2024 per the WHO. The bacterium's outer cell membrane is notoriously difficult to penetrate, rendering most antibiotics ineffective. Few drugs are currently available to treat the disease.
The New Techniques
A team led by the University of Massachusetts Amherst developed two complementary approaches. The first uses AI to screen millions of compounds for potential membrane-penetrating ability. The second, dubbed PAC-MAN, employs a molecular probe to directly measure how well candidate drugs cross the membrane. Together, they can rapidly identify promising drug candidates.
Impact and Next Steps
The techniques have already identified several compounds that show activity against Mtb in lab tests, according to the researchers. The team plans to refine these candidates and move toward animal studies. If successful, the methods could dramatically shorten the typical decade-long drug development timeline for TB.
What's Next
The researchers aim to begin animal testing of the most promising compounds within the next year. It remains unclear whether the techniques will yield a drug that is both safe and effective in humans, given the historical difficulty of targeting Mtb.
1 source
UMass team uses AI, PAC-MAN to speed TB drug discovery
