Estimating gene networks with cDNA microarray data using state-space models

Abstract

State-space model is used in this paper to analyze dynamics of gene expression profile data. State-space models describe dynamics that the observed measurements depend on some hidden state variables. Hidden state variables can capture effects that cannot be measured in gene expression profiling experiment, for example, genes that have not been included in the observation variables, levels of regulatory proteins or the effect of mRNA. System identification is achieved by EM algorithm that is based on the maximum likelihood method. We apply this method to a published yeast cell-cycle gene expression time-series data under the assumption that state variables and observation variables are generated by Gaussian white noise process, that produces the simplest and most reasonable model to explain of behaviors of the gene expression profiles. © Springer-Verlag Berlin Heidelberg 2005.