Systematic methodology for the development of mathematical models for biological processes

Abstract

Synthetic biology gives researchers the opportunity to rationally (re-)design cellular activities to achieve a desired function. The design of networks of pathways towards accomplishing this calls for the application of engineering principles, often using model-based tools. Success heavily depends on model reliability. Herein, we present a systematic methodology for developing predictive models comprising model formulation considerations, global sensitivity analysis, model reduction (for highly complex models or where experimental data are limited), optimal experimental design for parameter estimation, and predictive capability checking. Its efficacy and validity are demonstrated using an example from bioprocessing. This approach systematizes the process of developing reliable mathematical models at a minimum experimental cost, enabling in silico simulation and optimization. © 2013 Springer Science+Business Media, New York.