Time delay estimation in gene regulatory networks

Abstract

Gene regulation studies reveal unknown biological functions in disease progression. As more time-course datasets become available, interactions among the regulators and their associated target genes may better describe the evolution of gene regulatory networks. Currently in many research studies, interaction delay is not considered. Such delay is embedded in the network due to the intrinsic temporal process of gene expression. In this paper, a time delay regression model is developed to identify and predict time-dependent interactions. To estimate the model parameters, Average Square Difference Function and Least square estimation methods are applied. The time-course gene expression dataset in this paper was obtained for mice post-myocardial infarction. The simulation results show better performance of proposed method compared with no-delay and cross correlation-based methods.