MONET is a regulatory network inference algorithm based on Bayesian network learning, which enables genome-wide network inference from microarray expression data using parallel processing techniques with supercomputing resources. Therefore, without extracting subsets of genes from the expression data (e.g. DEG finding), researchers can perform network inference analysis using the whole data. The MONET algorithm, which stands for MOdularized NETwork learning, has a divide-and-conquer approach. It divides a whole gene set to overlapped modules considering biological annotations and expression data together, then infers a Bayesian network for each modules, and then integrates the learned subnetworks to a final, global network. This algorithm has been developed in a web service system with a parallel processing technique based on supercomputing center resources, and users also can use this service in Cytoscape's enviroments.