At the interface of biology, computer science and mathematics, a range of methods for the modelling, the analysis and the simulation of genetic regulatory networks have been developed. Such formal approaches allow the delineation of unambiguous descriptions of large and complex networks of interacting biological macromolecules, as well as predictions about their spatio-temporal behaviour, in normal or modified conditions. These theoretical methods can be subdivided into four broad categories depending on the mathematical formalism used: graphs, differential equations, stochastic models, and Boolean or generalised logical formalisms. Models further differ with respect to the level of granularity at which they describe regulatory networks, ranging from detailed molecular descriptions taking into account the stochastic effects arising from the small number of molecules of some components, to approximate models focusing on the global regulatory structure of a network. Several prokaryotic and eukaryotic regulatory networks have already been modelled and analysed, leading to new insights into the structure and functioning of these systems. Given the currently available genetic and molecular data, qualitative approaches, based on logical or differential equation models, seem particularly suitable. Still under development, these approaches and the corresponding computer tools should rapidly become indispensable for the study of the dynamical properties of normal and pathological biological networks, for the prediction of the effects of perturbations, and for the development of a new generation of therapeutic and agronomic tools.
|
