MODELING GENETIC REGULATORY NETWORKS: CONTINUOUS OR DISCRETE?

Abstract

Selecting an appropriate mathematical model to describe the dynamical behavior of a genetic regulatory network plays an important part in discovering gene regulatory mechanisms. Whereas fine-scale models can in principle provide a very accurate description of the real genetic regulatory system, one must be aware of the availability and quality of the data used to infer such models. Consequently, pragmatic considerations motivate the selection of a model possessing minimal complexity among those capable of capturing the level of real gene regulation being studied, particularly in relation to the prediction capability of the model. This paper compares fine-scale stochastic-differential-equation models with coarse-scale discrete models in the context of currently available data and with respect to their description of switch-like behavior among specific groups of genes.