Modeling & Process Optimization
Computational modeling and optimization are transforming modern biomanufacturing by enabling researchers to better understand, predict, and control complex cellular processes. As biologics become increasingly important in treating diseases ranging from cancer to infectious diseases, there is a growing need for efficient strategies that reduce development costs, accelerate process optimization, and improve product quality. By integrating mechanistic understanding of cellular metabolism with machine learning and advanced optimization techniques, digital biomanufacturing approaches provide a pathway toward more predictive, automated, and data-driven manufacturing platforms capable of producing safer, more affordable, and higher-quality therapeutics.
Our Research in modeling and optimization focuses on developing computational frameworks that describe and predict mammalian cell culture behavior across diverse bioprocess conditions. These efforts combine mechanistic metabolic models, hybrid modeling approaches, machine learning algorithms, and Bayesian optimization techniques to characterize cellular metabolism, optimize media formulations, and improve bioprocess performance. Applications include predicting cell growth and productivity, optimizing culture conditions, improving product quality attributes such as glycosylation, and enabling data-efficient process development. Through the integration of experimental data and computational intelligence, this work supports the development of next-generation digital biomanufacturing systems for therapeutic protein production.