Virtual Seminar Series on Bayesian Decision-making and Uncertainty

2025-05-05

The microbiome revolution has identified thousands of species of bacteria that deserve scientific investigation. Scientists cannot keep pace with the expanding tree of life, and most species of bacteria remain unstudied. AI and automation could accelerate scientific discovery by replacing humans with algorithms that mine the scientific literature and design new experiments. The least-studied bacteria would benefit the most from automated research, but, ironically, the lack of data makes it difficult to deploy autonomous agents to study these species. We present an automated science platform that maps microbial metabolism but requires no prior knowledge. Our AI agent – called BacterAI – learns by converting scientific questions into simple games that it plays with laboratory robots. The agent then distills its findings into logical rules that can be interpreted by human scientists. In less than two weeks, our system went from knowing nothing about a bacterium to communicating novel findings to another scientist.

Paul Jensen an assistant professor of biomedical engineering at the University of Michigan, where he mentors students in systems microbiology, artificial intelligence, and automated science. Paul trained as an engineer and microbiologist at the University of Minnesota, the University of Virginia, and Boston College. He is an ASQ Certified Quality Engineer and co-founder of the biotech company Cerillo, Inc. In 2020, Paul received the Stanley H. Pierce Faculty Award for developing empathetic student-faculty relationships.