Metabolic Flux Analysis of Escherichia coli Based on Kinetic Model and Genome-Scale Metabolic Network Model

The application of Genome-Scale Metabolic Network Models (GSMM) in fermentation optimization is hampered by challenges in differentiating viable from dead cells and parameter distortion induced by conventional detection methods. Using E. coli BL21(DE3) as the model organism, this study developed a flux analysis strategy that couples cell kinetics with GSMM. Key parameters were estimated using the gradient descent algorithm, thereby enabling precise prediction of viable cell concentration and glucose consumption dynamics. Integrating this with the Quadratic Programming-based parsimonious Flux Balance Analysis (QP-pFBA) algorithm, intracellular metabolic reaction fluxes were quantified. Results demonstrated that the model can effectively differentiate viable from dead cells; Batch D, adopting the gradient-increasing feeding strategy, achieved the maximum specific growth rate (μmax) of 0.6457, the highest among the four batches. Moreover, key metabolic reaction fluxes were highly correlated with the feeding strategy. This framework forgoes specialized, high-cost equipment and offers robust cross-strain/process adaptability, thereby greatly advancing GSMM utility. It provides a powerful tool for precise fermentation control and accelerates the shift toward data-driven biomanufacturing.