ySparsity Stability Causality Dynamic Bayesian networks

also at ple Available data sources include static steady state data and time course data obtained either for wild type ces, su y tand chrom Overall, the study of biological networks including modeling, anal-ysis, reconstruction and visualization aspects has become a key to-pic in bioinformatics and computational biology (for a review and recent trends see [4]). A number of learning tasks have been studied in the literature, based on the type of biological network under consideration. For example, in metabolic reaction networks, the focus has been on ctional influences cal interactions. ne in fun ct attests Two main types of data have been used to learn such net steady state data and time course data. steady state data tained from a long-term observation of gene expression, ass the system reaches an equilibrium state. For instance, multiple bio-logical replicates obtained at some late point in time provide such steady state data. Such data are usually obtained from microarray technologies, and provide a global view of the biological system under study in its natural state (wild type); however, their infor-mational content for network reconstruction purposes is in general limited and accurate network inference usually requires a ver