Using Coarse-Grained, Discrete Systems for Data-Driven Inference of Regulatory Gene Networks: Perspectives and Limitations for Reverse Engineering

This contribution gives an initial report of a new project exploring the perspectives and limits of reversely engineering regulatory gene networks from gene expression data. The availability of such data is currently increasing dramatically due to the microarray technology. However, inferring the underlying network from expression data is difficult. We address the reverse engineering problem by simulating the regulatory network and the process of extracting expression data. The simulated expression data thus obtained are then subjected to an algorithm for inferring regulatory networks. With this concept, we intend to assess the effects of statistical properties of the data extraction process on the suitability of the expression data for reverse engineering.