Inferring Alternative Splicing Gene Expression from Nextgen Sequencing Studies

• Activity Number: 623
• Type: Contributed
• Date/Time: Thursday, August 5, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): David Rossell*+
• Companies: IRB Barcelona
• Keywords: Nextgen sequencing ; Bayes ; mixture modeling ; RNAseq ; Model selection

Abstract

Next-generation sequencing (NGS) is a revolutionary tool for genomics which produces an overwhelming amount of data in a single experiment. Through the use of adequate statistical methods, NGS allows addressing biological questions at an unprecedented resolution. We focus on detecting and quantifying gene alternative splicing from paired-end RNAseq studies, an unfeasible problem with previous technology. The data can be viewed as arising from a mixture of probability distributions with an unknown number of components. We formulate a Bayesian model which takes full advantage of the structure of the data and permits incorporating (a priori) information available in public databases. The approach is computationally efficient and can be combined with standard model selection, model averaging and model checking procedures to deliver improved inference.