Analyzing Time Course Gene Expression Data with Biological and Technical Replicates to Estimate Gene Networks by State Space Models

In order to estimate accurate gene networks from time course gene expression data, replicated time course data are useful. However, existing methods do not clearly distinguish between biological and technical replicates, while these two kinds of replicates have different features. In this paper, we propose a statistical model based on state space models to use biologically and technically replicated time course data and show an algorithm to estimate a gene network that is a graphical representation of gene-gene regulation. To our knowledge, for estimating gene networks, the proposed model is the first model that can simultaneously use two types of replicated time course data. We show the effectiveness of the proposed method through the analysis of the microarray human T-cell data.