Construction of linear dynamic gene regulatory network based on feedforward neural network

Abstract

The purpose of analyzing gene network structure is to identify and understand some unknown related functions and the regulatory mechanisms at molecular level in organisms. Traditional model of the gene regulatory networks often lack an effective method of solving with gene expression profiling data because of high time and space complexity. In this study, a new model of gene regulatory network based on linear feedforward neural network is proposed. The new model combines the advantages of linear neural network including fast convergence, no existence of local minimum value, high precision and easy operation. It maps the linear neural network into complex network. Through statistics and comparison of network parameters, the differentially expressed genes related to the sample background can be identified. The numerical experiment in the latter part of the study verified the validity of the model.