Bayesian Functional Data Analysis over Dependent Regions and Its Application for Identification of Differentially Methylated Regions

TL;DR

This paper introduces a Bayesian functional data analysis method for long sequences with dependent windows, using smoothing splines and transition models, applied to identify methylation differences in cancer data.

Contribution

It develops a novel Bayesian approach combining smoothing splines and transition models to analyze dependent sequence windows for functional data analysis.

Findings

Effective in simulation studies for dependent sequences.

Successfully applied to identify differentially methylated regions in lung cancer data.

Provides a Bayesian framework for functional pattern estimation and group comparison.

Abstract

We consider a Bayesian functional data analysis for observations measured as extremely long sequences. Splitting the sequence into a number of small windows with manageable length, the windows may not be independent especially when they are neighboring to each other. We propose to utilize Bayesian smoothing splines to estimate individual functional patterns within each window and to establish transition models for parameters involved in each window to address the dependent structure between windows. The functional difference of groups of individuals at each window can be evaluated by Bayes Factor based on Markov Chain Monte Carlo samples in the analysis. In this paper, we examine the proposed method through simulation studies and apply it to identify differentially methylated genetic regions in TCGA lung adenocarcinoma data.