Predictable Whole Cell Kinetic Models with Default Enzyme Kinetics - mauriceling/mauriceling.github.io GitHub Wiki

Citation: Liew, NWJ, Mohamed-Khalid, N, Ng, FJY, Lim, TY, Abdul-Samathu, F, Ling, MHT. 2026. Predictable Whole-Cell Kinetic Models with Default Enzyme Kinetics. Medicon Medical Sciences 11(3): 30-35.

Link to [PDF].

Here is the permanent [PDF] and [dataset] links to my archive.

Whole-cell kinetic models (WCKMs), based on Michaelis-Menten kinetics, require accurate kinetic parameters; such as, turnover numbers (kcat), and Michaelis-Menten constants (Km); which are often unavailable. To address this, many ab initio constructions of WCKMs or conversion from genome-scale metabolic models relies on default kcat of 13.7 per second and Km of 130 micromolar, which are based on average enzyme kinetics. However, the predictability of WCKMs on default kcat and Km values is unclear. This study evaluates the impact of such defaults on WCKMs. Our results show that local sensitivity analysis (LSA) using one-factor-at-a-time revealed that kcat is the significantly more sensitive (p-values < 0.0269) than Km; and the Pearson’s correlation of time-course metabolite concentrations between using actual or default kcat values is 0.7166, suggesting substantial predictability despite using default kcat values. More importantly, fixing 5 % to 50% of the kcat values from default values to actual values (representing experimental verification of kcat values) show a monotonic improvement in predictability, where fixing 5% of reaction showed r = 0.717, and r = 0.826 at 50%. The findings imply that using default values provide a reasonable baseline but enzyme turnover numbers should be prioritised for experimental measurement during model construction as the benefit scales with the proportions of reactions validated.