NW: Biological Networks: Design Principles of Robust Information Processing

The computation ability of biochemical networks is striking when one considers that they function in biological environments where the amounts of network components fluctuate, the kinetics is stochastic, and sensitive interactions between computation modules are required. Several studies have examined the effect of these properties on cellular computing and robustness has been proposed to be an essential 'design principle' of biological networks. This project proposes a theoretical approach based on experimental data to investigate the underlying design principles leading to robust information processing for networks whose objective function is known. Well known examples for such networks are the chemotaxis pathway in bacteria, allowing the cells to sense gradients in chemo-ligand over four orders of magnitude and the pattern formed by the segment polarity network of the drosophila embryo that persists even under extreme variations in the values of interaction parameters. The novelty of this approach is to employ experimentally determined intercellular variations of the network components to reconstruct 'in silico' the set of smallest robust network structures which are capable to reproduce the experimentally observed sensitivity of the output signals. These minimal networks provide experimentally testable topologies and allow to distinguish the functional and noise-compensatory part of known network structures.