An integrated identifiability and metabolic flux analysis workflow for reducing and repairing dynamic CHO metabolic models

A reduced kinetically constrained dynamic flux balance analysis model of CHO perfusion culture shrank from 48 to 43 kinetic parameters while preserving performance across 23 extracellular species.

• Why it matters: Over-parameterized kinetic-metabolic models can match data yet produce unreliable parameters and misleading uncertainty, limiting their ability to transfer across media conditions.
• What they did: The workflow combined Fisher information matrix identifiability diagnostics, uncertainty propagation, and stoichiometric metabolic flux analysis to test transfer to amino acid-supplemented Medium C and to guide repair.
• The result: For Medium C, 79% of experimental points fell outside the 95% parameter-propagated interval, and MFA-guided repair reduced glutamate and lactate NRMSE by 64% and 36%, respectively.

AI summary of the abstract · may contain errors · not medical advice · read the paper.