In a significant breakthrough against antibiotic resistance, researchers at The University of Texas at Austin have leveraged artificial intelligence to develop a new antibiotic. Their study, featured in Nature Biomedical Engineering, represents a major advancement in creating safer, more effective treatments. The team utilized a large language model (LLM), similar to ChatGPT, to modify Protegrin-1, a potent antibiotic found in pigs that was previously too toxic for human use. By generating over 7,000 variants of Protegrin-1, they were able to identify modifications that retained its antibacterial properties while reducing toxicity. The LLM helped evaluate these variants for their ability to target bacterial membranes selectively, kill bacteria, and avoid damaging human red blood cells. This process led to the development of a refined antibiotic, called bacterially selective Protegrin-1.2 (bsPG-1.2). Preliminary trials in mice infected with multidrug-resistant bacteria showed that bsPG-1.2 significantly reduced bacterial levels within six hours. These promising findings suggest that bsPG-1.2 could advance to human trials. Claus Wilke, a professor of integrative biology and co-senior author of the study, emphasized the revolutionary impact of AI on drug development.
top of page
bottom of page
Comments