Principal components analysis based methodology to identify differentially expressed genes in time-course microarray data

BACKGROUND: Time-course microarray experiments are being increasingly used to characterize dynamic biological processes. In these experiments, the goal is to identify genes differentially expressed in time-course data, measured between different biological conditions. These differentially expressed genes can reveal the changes in biological process due to the change in condition which is essential to understand differences in dynamics. RESULTS: In this paper, we propose a novel method for finding differentially expressed genes in time-course data and across biological conditions (say C(1 )and C(2)). We model the expression at C(1 )using Principal Component Analysis and represent the expression profile of each gene as a linear combination of the dominant Principal Components (PCs). Then the expression data from C(2 )is projected on the developed PCA model and scores are extracted. The difference between the scores is evaluated using a hypothesis test to quantify the significance of differential expression. We evaluate the proposed method to understand differences in two case studies (1) the heat shock response of wild-type and HSF1 knockout mice, and (2) cell-cycle between wild-type and Fkh1/Fkh2 knockout Yeast strains. CONCLUSION: In both cases, the proposed method identified biologically significant genes.