Revisiting moment-closure methods with heterogeneous multiscale population models.

Area of Science:

• Multiscale modeling
• Chemical kinetics
• Systems biology

Background:

• Single-cell stochasticity drives population heterogeneity in genetically identical cells.
• This heterogeneity leads to diverse behaviors like varied growth and responses, complicating population dynamics.
• Accurate modeling of these multiscale processes is crucial but computationally expensive.

Purpose of the Study:

• To develop computationally efficient approximations for multiscale population models.
• To investigate moment-closure techniques for systems with single-cell stochasticity and population-level interactions.
• To address challenges in applying moment-closure methods to population models.

Main Methods:

• Order-reduction approximations based on distribution moments.
• Extension of single-cell moment-closure techniques to population models.
• Evaluation of closure methods for optimal control in a microbial consortium model.

Main Results:

• Simple moment-closure techniques can yield non-physical trajectories.
• Despite limitations, simple closures demonstrated superior accuracy, efficiency, and robustness in optimal bioproduction control.
• The study highlights the trade-offs and specific advantages of different closure methods.

Conclusions:

• Moment-closure approximations provide a viable approach to manage computational complexity in multiscale stochastic models.
• The choice of closure method significantly impacts model performance and the ability to solve optimization problems.
• This work offers insights into optimizing bioproduction through effective multiscale modeling and control strategies.