Abstract

AbstractRNA-Seq is a powerful tool for the study of alternative splicing and other forms of alternative isoform expression. Understanding the regulation of these processes requires comparisons between treatments, tissues or conditions. For the analysis of such experiments, we present DEXSeq, a statistical method to test for differential exon usage in RNA-Seq data. DEXSeq employs generalized linear models and offers good detection power and reliable control of false discoveries by taking biological variation into account. An implementation is available as an R/Bioconductor package.

Highlights

  • In higher eukaryotes, a single gene can give rise to a multitude of different transcripts by means of varying the usage of splice sites, transcription start sites and polyadenylation sites

  • I.e., instead of fitting one parameter βρCj for the effect of each condition ρ on the expression, we fit one parameter βiSj for each sample j, the gene expression variability is absorbed by the model parameters and we are only left with the exon usage variability

  • We performed the test for differential exon usage described in Section 2.2.5 for all counting bins that had at least 10 counts summed over all 7 samples

Read more

Summary

Background

A single gene can give rise to a multitude of different transcripts (isoforms) by means of varying the usage of splice sites, transcription start sites and polyadenylation sites. A notable instance where biological variation was accounted for in the statistical analysis is the work of Blekhman et al (2010). Their method relies on the availability of a moderate to large number of samples, and no software implementation was provided. In the Discussion (Section 4), we compare with competing methods, especially with the analysis provided by Brooks et al (2010) for their data (which is based on the method of Wang et al (2008)), and with the cuffdiff tool provided with the cufflinks software by Trapnell et al (2010)

Preparation
Model and Inference
A generalized linear model
Parameter fitting
Two noise components
Analysis of deviance
Additional covariates
Visualization
Application
Importance of modelling overdispersion
Analyses based on Fisher’s test
Heterogeneity of dispersions
Comparison with cuffdiff
Comparing exon or isoform usage
Implementation
Conclusion
The negative binomial distribution from a gamma-Poisson hierarchical model
Balancing
Details on the Cox-Reid dispersion estimation
Variance stabilizing transformation
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.