Identifying and Clustering Stable and Dynamic CpG Sites via Bayesian Analysis

• Activity Number: 477 - SPEED: Bayesian Methods and Applications in the Life and Social Sciences
• Type: Contributed
• Date/Time: Wednesday, August 1, 2018 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Luhang Han* and Hongmei Zhang and Ebenezer Olusegun George and Wilfried Karmaus and Hasan Arshad and John Holloway
• Companies: University of Memphis and University of Memphis and University of Memphis and University of Memphis and University of Southampton and University of Southampton
• Keywords: Bayesian inference; Clustering; Epigenetic; Adolescence

Abstract

This piece of research work was motivated by a genome-wide epigenetic study, where epigenetic changes from pre- to post-adolescence was of interest due to their potential connection to allergic diseases. We focused on one type of epigenetic assessment, DNA methylation at a certain number of CpG sites (i.e., measurements for a certain number of variables). To efficiently and effectively characterize DNA methylation at different CpG sites, we developed a Bayesian two-stage clustering method to 1) determine whether DNA methylation at a CpG site was stable over time, and 2) assign each unstable CpG site into a specific cluster based on temporal trend of DNA methylation at that site. Simulations were used to demonstrate and assess the developed method. We then applied the approach to a real data set composed of DNA methylation at 2,000 CpG sites measured at 10 and 18 years, respectively, for each of 325 subjects.