University of Florida researchers are on a mission to transform biotechnology and improve the way genetic disorders are treated, with help from generative artificial intelligence, or AI, and a U.S. Department of Defense grant.
Wenjun Xie, Ph.D., an assistant professor of medicinal chemistry in the UF College of Pharmacy, has received $100,000 in funding from the Defense Advanced Research Projects Agency, or DARPA, which is responsible for the development of emerging technologies for use by the military. Through this funding opportunity, DARPA solicited innovative ideas demonstrating AI and machine learning’s capabilities in designing proteins and other biomolecules to improve health outcomes.
“DARPA funds research with the potential for transformative impact, and our work aligns well with this mission,” said Xie, the principal investigator for the project, who is working alongside Jing Pan, Ph.D., an assistant professor of mechanical and aerospace engineering in the UF Herbert Wertheim College of Engineering. “Receiving this support will accelerate our research and help us realize some long-term goals around enzyme engineering in the near future.”
Mapping the path forward, with generative AI
“If an enzyme undergoes a mutation, it can lead to a genetic disease. By engineering enzymes, we can use them directly as therapeutics to treat these conditions. Additionally, we can design small molecules that target enzymes to treat various diseases,” Xie said.
Generative AI serves as a critical tool in Xie’s arsenal. Unlike traditional AI, which follows fixed rules and algorithms, generative AI learns from vast amounts of data. In research published in the Journal of the American Chemical Society in January, Xie and his collaborators demonstrated generative AI’s ability to analyze enzyme sequences, leading to the successful engineering of enzymes.
“We’re all using ChatGPT, right? ChatGPT, powered by generative AI, learns from human language and understands its semantics,” Xie said. “In our research, we’re applying a similar approach — not to human language, but to the language of proteins. Here, semantics translates to the physical chemistry of enzymes.”
Xie’s work in designing highly efficient enzymes — proteins that help speed the body’s metabolism — has the potential to transform the treatment of genetic disorders, though its uses expand far beyond therapeutics.
Transforming the future of biotechnology
Despite generative AI’s tremendous power, its limitations hinder the ability to design novel enzyme activity. Xie’s research will combine generative AI with molecular simulations to enable scientists to design efficient enzymes.
“We want to design novel enzyme activity that hasn’t been seen in nature. To do so, we need to understand the principles underlying enzyme catalysis, especially physical chemistry principles,” Xie said. “Generative AI gives us a unique viewpoint so that we can systematically understand these principles.”
Xie said his research has found a home within the UF Department of Medicinal Chemistry, where computational scientists, life scientists and chemists work side by side in areas like AI-driven drug discovery, enzymology and enzyme engineering, biochemistry and bioinformatics, and natural product biosynthesis.
As one of nine new faculty members focused on AI within the college — and one of more than 100 across UF — Xie’s role supports the university’s mission to lead in AI-driven education and research.
“Our goal is to design an efficient enzyme to catalyze any given chemical reaction,” Xie said. “If we can realize this goal, we can transform how we work in biotechnology.”