Efficient dimensional and computational reduction of lactate dehydrogenase (LDH) system by integrating QSSA, CSP,...

This study investigates the application of analytical and computational techniques to simplify enzyme-catalyzed multi-substrate reaction systems, emphasizing dimensional reduction strategies for complex kinetic models. By integrating the Michaelis-Menten framework with methodologies such as Quasi-Steady-State Approximation (QSSA), Computational Singular Perturbation (CSP), and sensitivity analysis, we validate low-dimensional approximations that capture the temporal dynamics of fast and slow species in enzymatic interactions. A key innovation is the separate evaluation of the Slow Invariant Manifold (SIM) for reduced species, diverging from conventional approaches that analyze all species collectively. Utilizing MATLAB's Simbiology and Simulink toolboxes, we demonstrate enhanced computational efficiency and dynamic visualization of reaction mechanisms, enabling robust, geometrically interpretable comparisons of species behavior. The results underscore the practical utility of these techniques in biochemical and industrial contexts, particularly for optimizing dissipative systems in chemical kinetics. This work advances model reduction methodologies by providing refined insights into enzyme-substrate dynamics and establishing a computational framework for future extensions to nonlinear and stochastic reaction networks.