A Bayesian Multi-Scale Poisson Model for Detecting Differences Between Multiple Groups in High-Throughput Sequencing Data and Its Application to Small Sample Sizes

• Activity Number: 20
• Type: Topic Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Heejung Shim*+ and Zhengrong Xing and Ester Pantaleo and Matthew Stephens
• Companies: University of Chicago and University of Chicago and University of Chicago and University of Chicago
• Keywords: high-throughput sequencing ; multi-scale Poisson ; RNA-seq ; DNase-seq ; ChIP-seq ; wavelet

Abstract

Identification of differences between multiple groups in molecular and cellular phenotypes measured by high-throughput sequencing assays is frequently encountered in genomics applications. For example, common problems include detecting differential gene expression between multiple conditions using RNA-seq and detecting differences in transcription factor binding/chromatin accessibility across tissues using DNase-seq or ChIP-seq. Motivated by our previous wavelet-based approach to genetic association analysis of functional phenotypes that better exploits high-resolution information from high-throughput sequencing assays, here we present statistical methods that model the count nature of the sequence data directly using multi-scale models. Specifically, our methods consider the data as an inhomogeneous Poisson process and test for differences in underlying intensities using a Bayesian multi-scale model. Unlike the previous wavelet-based approach, the proposed methods can provide the power to detect differences with small sample sizes. We illustrate the proposed methods on ENCODE DNase-seq data from multiple tissues.