A systems approach to discovering signaling and regulatory pathways--or, how to digest large interaction networks into relevant pieces.

In the post-genomic era, the first step in any study of protein function is a homology search against the complete genome sequence of the organism of interest. By analogy, if we also wish to elucidate the cadre of signaling and regulatory pathways in the cell, an extremely powerful first step is to construct a complete network of protein-protein and transcriptional interactions and then search through this network to identify critical pathways in a top-down fashion. Like genomic sequence, the molecular interaction network provides a broad foundation for more directed studies to follow. We illustrate this strategy using a large network of 12,232 interactions in yeast. A variety of applications are discussed, including screening the network to identify pathways responsible for gene expression changes observed in galactose-induced cells, and identifying groups of interacting proteins that are essential (by phenotypic assay) for the cellular response to DNA damage.