Additional material to DESeq2 paper

Additional material for

Moderated estimation of fold change and dispersion for RNA-Seq data with DESeq2

by Michael I Love, Wolfgang Huber, and Simon Anders

This page provides access to additional material for our paper on DESeq2, namely to all R code and scripts used to produce the figures and benchmark results in the paper.

Paper package

The R package DESeq2paper contains all code, data and vignettes and is available as a tarball.

DESeq2paper_1.3.tar.gz (current version)

The package for previous version of the paper is provided for archival purposes:

DESeq2paper_1.2.tar.gz (outdated)
DESeq2paper_1.1.tar.gz (outdated)
DESeq2paper_1.0.tar.gz (outdated)

DESeq2 package

The R package DESeq2 can be obtained from the Bioconductor website.

DESeq2

Vignettes

The following vignettes are all included in the package but can be downloaded here separately

Shrinkage of dispersion estimates - Rnw
Shrinkage of logarithmic fold changes - Rnw
Stability of shrunken logarithmic fold changes - Rnw
Testing logarithmic fold changes at a nonzero threshold - Rnw
Count outlier detection using Cook’s distance - Rnw
Regularized logarithm for sample clustering - Rnw
Assessment of DESeq2 performance through simulation - Rnw
Specificity using Pickrell et al dataset - Rnw
Sensitivity and precision using the Bottomly et al. dataset - Rnw
Theoretical and sample variance of logarithmic dispersion estimates - Rnw
Assessing dispersion estimation and shrinkage through simulation - Rnw
Estimation of mean dispersion pairs from Pickrell et al dataset - Rnw
Marginal independence of the test statistic of the filter statistic - Rnw
Observed log fold changes depend on the mean for simulated counts - Rnw

Scripts

For the scripts used for the benchmarking, see the directory /inst/script/ in the tarball, or the files here:

Scripts

R data objects

meanDispPairs.RData
bottomly_sumexp.RData
pickrell_sumexp.RData