Clustering Gene Expression Data Based on Predicted Differential Effects of GV Interaction

Microarray has become a popular biotechnology in biological and medical research.However, systematic and stochastic variabilities in microarray data are expectedand unavoidable, resulting in the problem that the raw measurements have in-herent "noise" within microarray experiments. Currently, logarithmic ratios areusually analyzed by various clustering methods directly, which may introduce biasinterpretation in identifying groups of genes or samples. In this paper, a statisticalmethod based on mixed model approaches was proposed for microarray data clus-ter analysis. The underlying rationale of this method is to partition the observedtotal gene expression level into various variations caused by different factors usingan ANOVA model, and to predict the differential effects of GV (gene by variety)interaction using the adjusted unbiased prediction (AUP) method. The predictedGV interaction effects can then be used as the inputs of cluster analysis. We illus-trated the application of our method with a gene expression dataset and elucidatedthe utility of our approach using an external validation.