Scaling microbial fermentation from the laboratory to industrial production presents significant challenges, particularly in maintaining stability during continuous operation. Using a β-carotene-producing Yarrowia lipolytica strain as a model, the Tang Lab and collaborators investigated the key factors contributing to titer loss, focusing on bioreactor modes, carbon sources, oxygen availability, and media composition. Their findings showed that fermentation mode and oxygen conditions had the greatest impact on strain stability. Notably, oil-based carbon sources significantly enhanced both titer and production longevity compared to glucose-based media. To uncover the mechanisms driving titer loss, the team employed omics analyses and kinetic modeling, revealing contributions from subpopulation dynamics and spontaneous mutations. In parallel, large language models and machine learning tools were leveraged to accelerate strain development and fermentation optimization by extracting insights from vast scientific literature. This research has also catalyzed several new projects in innovative biomanufacturing, integrating synthetic biology, AI, and process engineering approaches.
Professor Tang did his PhD at the University of Washington and postdoc at Lawrence Berkeley National Lab. He joined the School of Engineering faculty at Washington University in 2008. His research focuses on applied environmental microbiology, metabolic modeling and bioprocess engineering. Professor Tang is serving on several Editorial Boards, namely Biotech Advance, Biotechnology for Biofuel, Current Opinion in Biotechnology, and Process Biochemistry. He has received awards including NSF CAREER (2010), Ralph E. Powe Junior Faculty Enhancement (2010), and NSF Transition (2023). At Washington University, he taught Process Control, Metabolic Engineering, Fluid Mechanics, Bioprocess Engineering, and Large Language Model for Bioengineering. He earned a Department Chair’s Award for Outstanding Teaching in 2013.