An R package for Differential Expression Analysis. Given count data from two experimental conditions, denoiSeq helps one determine which transcripts are differentially expressed across the two conditions using Bayesian inference of the parameters of a bottom-up model for PCR amplification developed in “Chromatin conformation governs T cell receptor J beta gene segment usage”, by Ndifon et al.
To use the package, one needs to create a readsData
object and invoke the denoiseq
function on it. The results are obtained from the return value of denoiseq using the results
function which then computes the test statistic used in differential analysis.
RD <- new("readsData", counts = ERCC) #creating the readsData object
steps <- 3000 #steps for MCMC
BI <- denoiseq(RD, steps) #invoking denoiseq on the readsData object
rez <- results(BI,steps) #computing the test statistic
This package can be istalled from CRAN using install.packages(“denoiSeq”) or from github using devtools::install_github(“buriom/denoiSeq”).