sSeq-package: Shrinkage estimation of dispersion in Negative Binomial...

In sSeq: Shrinkage estimation of dispersion in Negative Binomial models for RNA-seq experiments with small sample size

Description

This package is to discover the genes that differentially expressed between two conditions based on RNA-seq experiments. Gene expression is measured in counts of transcripts and modeled with the Negative Binomial (NB) distribution using a shrinkage approach for dispersion estimation. The method of moment (MM) estimates for dispersion are simply shrunk toward a target, which minimizes the average squared difference between the shrinkage estimates and the initial estimates. The exact per-gene probability under the NB model is calculated, and used to test the hypothesis that the expected expression of a gene in two conditions are not different.

Details

Package:	sSeq
Type:	Package
Version:	1.0
Date:	2013-02-25

Author(s)

Danni Yu <[e-mail]>, Wolfgang Huber <[e-mail]> and Olga Vitek <[e-mail]>

References

Shrinkage estimation of dispersion in Negative Binomial models for RNA-seq experiments with small sample size

Examples

#load a simulated data that includes a count table
data("countsTable")

#calculate the p-values using the shrinkage approach.
conds <- c("A", "B")
resJS <- nbTestSH( countsTable, conds, "A", "B")

sSeq documentation built on Nov. 8, 2020, 5:52 p.m.