Researchers at Texas A&M University have developed new artificial intelligence tools to improve tuberculosis drug discovery. Led by James Sacchettini, the team built the AI system to help scientists narrow down thousands of potential compounds after initial screening. "We might get thousands of compounds from a screen and then have to decide which one we are going to work on," Sacchettini explained. The new tool also organizes years of collaborative research into a searchable database, helping researchers make better decisions about which drugs to pursue.
Tuberculosis remains the world's deadliest infectious disease according to the World Health Organization. The bacterial illness has affected humans for thousands of years and typically requires months of treatment. Drug-resistant strains and cases involving HIV co-infection take even longer to treat. The disease disproportionately affects lower-income regions where limited healthcare infrastructure makes control particularly challenging. A 1990s outbreak in New York City, particularly on Rikers Island, demonstrated that TB remained a serious threat.
Developing new TB treatments faces significant scientific obstacles. The tuberculosis bacteria possess a thick, waxy coating that prevents most drugs from reaching their cellular targets. Additionally, the bacteria grow very slowly compared to other pathogens, meaning experiments can take months rather than weeks. This slow growth rate has historically made the drug discovery pipeline lengthy. AI technology offers the potential to accelerate this traditionally slow process.