Online Program Home
My Program

Abstract Details

Activity Number: 485 - Bayesian Latent Variable Methods for Life Sciences
Type: Contributed
Date/Time: Wednesday, August 1, 2018 : 8:30 AM to 10:20 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #329323 Presentation
Title: Bayesian Latent Class Models for Identifying Biomarkers in Circadian Patterns
Author(s): Sung Duk Kim* and Paul S Albert
Companies: National Cancer Institute and National Cancer Institute
Keywords: Circadian pattern; Latent class model; Metabolite data; Random effect

Researchers in biology and medicine have increasingly focused on characterizing circadian rhythms and their potential impact on disease. Epidemiological studies from recent decades have supported a unique role for circadian rhythm in metabolism. As evidenced by individuals working night or rotating shifts, but also by rodent models of circadian arrhythmia, disruption of the circadian cycle is strongly associated with metabolic imbalance. In this paper, we develop new statistical methodology for identifying biomarkers in circadian patterns using metabolites that are observed over time in a population of individuals. We develop latent class approach to incorporate circadian patterns with a harmonic representation that allows for individual variation in the phase-shift across subjects. Markov chain Monte Carlo sampling is used to carry out Bayesian posterior computation. Several variations of the proposed models are considered and compared using the deviance information criterion. We illustrate this methodology with high-dimensional metabolite data

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program