Model-Guided Strain Engineering: Bridges between Design and Learning in Bioprocess Development

Summary

Industrial biotechnology focuses on the use of microorganisms, referred to as cell factories, to produce bio-based chemicals as an alternative to petroleum-based production contributing to the fight against climate change. However, designing these cell factories is difficult since we do not completely understand how a cell works or how the desired production and the environmental conditions affect the functioning of the cell. The complex interplay among the numerous factors that affect the performance of bioprocesses prevents accurate predictions when genetic or environmental variables are perturbed. This results in long developmental times that hinder the market implementation of biotechnological products. In this thesis, mathematical modeling is used to iteratively improve the performance of cell factories, focusing on experimental design and leveraging the information gained through experimentation. Knowledge-based models including kinetic models and genome-scale metabolic modeling are combined with the analysis of omic data and the use of statistical design of experiments and machine learning to accelerate cell factory and bioprocess design.