Identifying Heterogenous Transgenerational DNA Methylation Sites via Clustering in Beta Regression

• Activity Number: 84
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Shengtong Han*+ and Hongmei Zhang and Gabrielle A. Lockett and Nandini Mukherjee and John W. Holloway and Wilfried Karmaus
• Companies: and University of Memphis and University of Southampton and University of Memphis and University of Southampton and University of Memphis
• Keywords: DNA methylation transmission ; clustering ; EM algorithm

Abstract

This paper explores the DNA methylation transmission issue via clustering approach. Beta regression is employed to model the transmission pattern from parents to their offsprings. Given fixed number of clusters, EM algorithm is utilized to optimize the parameters and cluster assignments in each iteration. BIC is used to determine the final number of clusters and consequently the final clustering assignments is obtained as well. The efficiency of the proposed method is tested by simulated data sets. A real data set is analyzed and new findings are obtained, which may provide more meaningful insight on methylation transmission issue.