Semi-mechanistic kinetic (i.e., dynamic) models based on first principles are particularly relevant in biology, as they can explain and predict functional behavior that arises from varying concentrations of the cellular components over time. Here, we describe a computational tuning framework to facilitate both the selection of kinetic parameters for these models and its estimation from experimental data. On the one hand, the tuning framework uses multi-objective optimization to generate a model-based set of guidelines for the selection of the kinetic parameters. These parameter values are the required ones to provide a biological system with desired behavior, while fulfilling the design criteria encoded in the optimization problem itself. On the other hand, this framework can also be used to estimate the parameter values of biological systems from experimental data, once the optimization objectives had been defined appropriately. The methodology gives accurate identification results, as it provides clear orientation on the effect of the parameter values over the system's behavior even under different experimental scenarios. It is particularly useful for easily combining time-course-averaged data and steady-state distribution data. This protocol also addresses aspects related to the appropriate description of the kinetic models and the settings of the software tools. Therefore, it supplies for hands-on testing to evaluate the validity of the underlying technical assumptions of the biological kinetic models.
Keywords: Kinetic models; Multi-objective optimization; Parameter estimation; Parameter selection.
© 2022. Springer Science+Business Media, LLC, part of Springer Nature.