System for automatically inferring a genetic network from expression profiles

A system is constructed to automatically infer a genetic network by application of graphical Gaussian modeling to the expression profile data. Our system is composed of two parts: one part is automatic determination of cluster boundaries of profiles in hierarchical clustering, and another part is inference of a genetic network by application of graphical Gaussian modeling to the clustered profiles. Since thousands of or tens of thousands of gene expression profiles are measured under only one hundred conditions, the profiles naturally show some similar patterns. Therefore, a preprocessing for systematically clustering the profiles is prerequisite to infer the relationship between the genes. For this purpose, a method for automatic determination of cluster boundaries is newly developed without any biological knowledge and any additional analyses. Then, the profiles for each cluster are analyzed by graphical Gaussian modeling to infer the relationship between the clusters. Thus, our system automatically provides a graph between clusters only by input the profile data. The performance of the present system is validated by 2467 profiles from yeast genes. The clusters and the genetic network obtained by our system are discussed in terms of the gene function and the known regulatory relationship between genes.