An information-theoretic algorithm to data-driven genetic pathway interaction network reconstruction of dynamic systems

High-throughput technologies for biomolecular measurements and computational methods generate vast amounts of data. Quantitative analysis of such datasets is a key goal of systems biology that aims to understand the underlying processes and structures of complex biological systems. While several techniques have been developed to identify biological networks from steady-state data, only a few of them work well for dynamic networks. Development of computational algorithms to reconstruct biological networks from time-series measurements remains an important challenge in bioinformatics and systems biology. We propose an information-theoretic algorithm to reconstruct networks from microarray time-course data by identifying the topology of functional sub-networks. We employ our approach to reconstruct genetic pathway interaction network of yeast cell-cycle.