Modeling and identification of biological systems with emphasis on osmoregulation in yeast

Doctoral thesis, 2005

This thesis deals with two topics in the area of systems biology. The first topic, model identification, concerns the problem of automatically identifying a mathematical model of a biochemical system from experimental data. We present algorithms for model selection and parameter estimation that identify both the structure and the parameters of a differential equation model from experimental data. The algorithms are designed to handle problems of realistic size, where reactions can be non-linear in the parameters and where data can be sparse and noisy. To achieve computational efficiency, parameters are estimated for one equation at a time, giving a fast and accurate parameter estimation algorithm compared to other algorithms in the literature. The model selection is done with an efficient heuristic search algorithm, where the structure is built incrementally. The main strengths of our algorithms are that a complete model, and not only a structure, is identified, and that they are considerably faster compared to other identification algorithms. The second topic concerns mathematical modeling of osmoregulation in the yeast \emph{Saccharomyces cerevisiae}. This system involves the biophysical and biochemical responses of a cell when it is exposed to an osmotic shock. We present two different differential equation models based on experimental data of this system. The first model is a detailed model taking into account an extensive amount of molecular detail, while the second is a simple model with less detail. We demonstrate that both models agree well with experimental data on wild-type cells. Moreover, the models predict the behavior of other genetically modified strains and input signals.