A Hierarchical Approach to Analyze RNA-Seq Data with Length Bias

• Activity Number: 121
• Type: Topic Contributed
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Patrick Harrington*+ and Lynn Kuo
• Companies: University of Connecticut and University of Connecticut
• Keywords: 454 ; RNA-Sequence ; Length Bias ; Hierarchical

Abstract

Next-generation sequencing is being used to identify genes which are involved in various processes. This new technology analyzes RNA-seq data. Various assumptions have been made in order to analyze this data. Since it is count data, obvious parametric assumptions include Poisson, Negative Binomial, and hierarchical Negative Binomial. Beyond the typical struggles data may present, it has been shown empirically that differentially expressed transcripts which are of a longer length are more likely to be identified than their shorter counterparts. While methods have been suggested to correct for this bias, we suggest a hierarchical Negative Binomial model which aims to identify DE genes and address this length bias.