Multiscale Simulation of Large Numbers of Interacting Biological Cells

Adrien Coulier
Division of Scientific Computing
Department of Information Technology
Uppsala University
Uppsala


Abstract:

In the field of Systems Biology, noise plays an important role due to the low copy number of molecules in living cells. Stochastic models are therefore an unavoidable tool to study this phenomenon. Simulating stochastic models, however, remains computationally expensive, especially when considering large populations of cells.

There is therefore an increasing need for model simplification methods. Cells, however, remain incredibly complex organisms who are far from being completely understood, which makes model simplification particularly challenging. As a matter of fact, understanding the origins of intrinsic noise in cells is a critical question to address.

The goal of this research project is then twofold: first we want to quantify the effects of various approximation techniques on the results of a simulation. Second, we want to develop adaptive methods that will choose which level of approximation should be used in order to optimize the accuracy and the computational cost of a simulation.