Discovering dynamic regulatory pathway by applying an auto regressive model to time series DNA microarray data

In this paper we propose a novel method to extract dynamic regulatory pathways from time-series DNA microarray data. To this aim, first a specialized clustering technique is applied that utilizes the available heuristic information about the biological system to form the clusters. Then, an auto regressive (AR) model is applied to model the interactions among all genes and to predict the gene expressions for the next time steps. We tested the proposed method on the eukaryotic cell cycle data. The results indicate that the proposed method can successfully predict the dynamic pathway involved in this biological process.