Creates gene networks for a target sample in both dense and sparse region. NetworkProfiler is a statistical approach based on varying coefficients and a kernel-based L1-type regularization method. The algorithm aims to generate patient-specific, modulator-dependent gene regulatory networks. It is able to model a sample according to the grouped neighborhood around it and to produce its own gene regulatory network.
Faculty of Global and Science Studies, Yamaguchi University, Japan; Graduate School of Medicine, Nagoya University, Nagoya, Japan; Health Intelligence Center, Institute of Medical Science, University of Tokyo, Tokyo, Japan; Human Genome Center, Institute of Medical Science, University of Tokyo, Tokyo, Japan
NetworkProfiler funding source(s)
Supported by MEXT as Priority Issue on Post-K computer (hp150265, hp160219, hp1702217) and Innovative Area (15H05192).