Hello,

Based on other threads on this forum, I was planning to use edgeR:::treatDGE() to select differential open chromatin peaks while taking into account fold change.

I installed edgeR using biocLite and library(edgeR), and have been able to use many common functions from edgeR (i.e. DGEList, estimateGLM*Disp, glmFit and glmLRT). However, R cannot find treatDGE. The lengthy results of ls(package:edgeR) and ls(getNamespace("edgeR"), all.names=T) both did not include "treatDGE".

Did I install edgeR incorrectly, was the function deprecated, or is there another explanation?

Thanks, Eric

> sessionInfo()

R version 3.0.2 (2013-09-25)
Platform: x86_64-pc-linux-gnu (64-bit)

locale:
[1] C

attached base packages:
[1] grid      stats     graphics  grDevices utils     datasets  methods  
[8] base     

other attached packages:
 [1] BiocInstaller_1.10.4 edgeR_3.4.2          limma_3.16.7        
 [4] ctc_1.34.0           amap_0.8-14          cba_0.2-14          
 [7] proxy_0.4-14         gplots_2.11.3        MASS_7.3-29         
[10] KernSmooth_2.23-10   caTools_1.14         gdata_2.13.2        
[13] gtools_3.2.1        

loaded via a namespace (and not attached):
[1] bitops_1.0-6 tools_3.0.2 

You are running a quite old version of R (from 2013-09-25, note your sessionInfo output), which means you are getting the correspondingly old version of Bioconductor. You should update to the latest version to get new features.

In addition to Ryan's comment, treatDGE() is almost surely not what you want. That function is intended for RNA-Seq analysis of gene expression.

If I assume you are doing something like DNAse H1 digestion of your DNA to find regions of open chromatin, then (to me, at least) an analogous method is ChIP-Seq, where you are pulling down unknown regions of the genome, and then are A.) trying to figure out which regions you have pulled down, and B.) trying to see if you are pulling down different regions in different samples (or different amounts of a particular region in different samples). The only difference to me is that you are 'pulling down' regions of DNA with DNAse H1, rather than IP.

If that assumption holds, then edgeR is not the tool for you. Instead, I would look at ChIP-Seq type tools: http://bioconductor.org/packages/release/BiocViews.html#___ChIPSeq

If you like the software coming out of Gordon Smyth's lab, then you might take a look at csaw.