Abstract

We have modeled genetic regulatory networks in the framework of continuoustime recurrent neural networks. A method for determining the parameters of such networks, given expression level time series data, is introduced and evaluated using artificial data. The method is also applied to a set of actual expression data from the development of rat central nervous system. 1 Introduction In recent years, the available amount of gene expression data has been increasing at a rapid rate, and there now exist large data sets with expression level measurements from different tissue types and different organisms. The existence of time series data, i.e. measurements of the time variation of the expression levels of a number of genes, raises the possibility of determining the regulatory interactions between genes. Recently, there have been many efforts to model genetic regulatory networks and to determine the interactions within them, using several different mathematical frameworks (see e.g. Kauffma...

Citations