A Hierarchical Bayesian Model for Estimating and Inferring Differential Isoform Expression for Multi-sample RNA-Seq Data

Abstract

RNA-Seq has drastically changed our ways of studying transcriptomes in providing more precise estimates of gene expression, including isoform-specific expression. Most of the available methods for RNA-Seq data focus on one sample at a time. We present in this paper a Poisson-Gamma hierarchical model for multi-sample RNA-Seq data analysis in order to simultaneously estimate isoform-specific expression and to identify differentially expressed isoforms. Our model has the advantage of borrowing information across all samples in estimating expression levels, which can improve the estimates drastically, particularly for low abundance isoforms. Furthermore, our hierarchical model has the ability to account for overdispersion in the data and also can incorporate sample-specific covariates in the underlying model, which facilitates the isoform-specific differential expression analysis. Simulation studies demonstrated that this Bayesian multi-sample approach can lead to more precise estimates of isoform-specific expression and higher power to detect differential expression by borrowing information across all samples than single-sample analysis, especially for isoforms of low abundance. We further illustrated our methods using the RNA-Seq data of 10 Yoruban and 10 Caucasian individuals.