High-Dimensional Linear State Space Models for Dynamic Gene Regulatory Networks

• Activity Number: 570
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Iris Chen*+ and Hulin Wu
• Companies: University of Rochester and University of Rochester
• Keywords: State Space Model ; dynamic Gene Regulatory Network ; adaptive Lasso

Abstract

Gene regulation has been extensively studied for different biological systems. Expression profiles from time-course experiments along with appropriate mathematical models will allow us to identify dynamic regulatory networks. However, the challenges fall on the high-dimensional nature of such data without a surprise. We propose a high-dimensional linear State Space Model (SSM) with a new Expectation-Regularization-Maximization (ERM) algorithm to construct the dynamic gene regulatory network (GRN). System noise and measurement noise can be separately specified through SSMs. In order to deal with the problem of high-dimensional parameter space in the SSMs, the proposed new ERM algorithm employs the idea of the adaptive LASSO-based variable selection method so that the sparsity property of gene regulatory networks can be preserved. The simulation studies are performed to evaluate the proposed ERM algorithm for variable selection. The proposed method is applied to identify the dynamic GRN from time-curse gene expression data of yeast cell cycle experiments.