Finding the Circadian Clocksin Genes: An Application of Dirichlet Process Mixture Model and Spectral Analysis

• Activity Number: 495
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Yan Ren*+ and Christian I. Hong and Seongho Song
• Companies: University of Cincinnati and University of Cincinnati and University of Cincinnati
• Keywords: Dirichlet process mixture model ; Bayesian ; spectral analysis ; fast Fourier transform ; microarray data ; gene circadian clock

Abstract

In recent decades, more and more biological researchers are interested in studying genes. One important property of genes is that some genes have oscillating expressions over time. Considering the gene expression as an oscillation, researchers are interested to know the period, amplitude, and phase of the oscillation. However, microarray data of gene expressions shows dramatic difference among genes. So we propose to first cluster the genes by a Bayesian Dirichlet process mixture model, and then apply the spectral analysis to the cluster means by implementing fast Fourier transform to estimate the properties of the dominating spectrum. We apply the method to the microarray data described in Wubei Dong, et al (2008).