GRASP: a computational platform for building kinetic models of cellular metabolism

Abstract

Summary Kinetic models of metabolism are crucial to understand the inner workings of cell metabolism. By taking into account enzyme regulation, detailed kinetic models can provide accurate predictions of metabolic fluxes. Comprehensive consideration of kinetic regulation requires highly parameterized non-linear models, which are challenging to build and fit using available modelling tools. Here, we present a computational package implementing the GRASP framework for building detailed kinetic models of cellular metabolism. By defining the mechanisms of enzyme regulation and a reference state described by reaction fluxes and their corresponding Gibbs free energy ranges, GRASP can efficiently sample an arbitrarily large population of thermodynamically feasible kinetic models. If additional experimental data are available (fluxes, enzyme and metabolite concentrations), these can be integrated to generate models that closely reproduce these observations using an Approximate Bayesian Computation fitting framework. Within the same framework, model selection tasks can be readily performed. Availability and implementation GRASP is implemented as an open-source package in the MATLAB environment. The software runs in Windows, macOS, and Linux, is documented (graspk.readthedocs.io) and unit-tested. GRASP is freely available at github.com/biosustain/GRASP. Supplementary information Supplementary data are available at Bioinformatics Advances online.

Bayesian Regression Facilitates Quantitative Modeling of Cell Metabolism

“…We used the model of enzyme saturation from Liebermeister et al and the generalized Monod–Wyman–Changeux model of Allosteric regulation introduced in − and used more recently in Matos et al To capture the effect on enzyme activity of coupled phosphorylation and dephosphorylation processes, we developed a new mathematical model inspired by the generalized MWC model of allosteric regulation. Full details of all mathematical aspects of Maud’s kinetic model can be found in Supporting Information Section 2.…”