EdgeR condition-specific dispersion

tfwillems · 10-03-2012, 12:46 PM

I'm dealing with a factorial RNA-seq data set in which cells have been stimulated with various combinations of extra-cellular cues. As such, I was interested in applying the GLM framework in edgeR to assess the contribution of each extra-cellular cue to the differential expression of certain genes. My concern, however, is that both the expression level and the dispersion of each gene varies greatly with the combination of cues. EdgeR doesn't seem to estimate condition-specific dispersion but rather one dispersion per gene (if the tagwise options is used). My question is therefore two-fold:
1) Does it make sense to want to estimate condition-specific dispersions?
2) Is there a way to modify the edgeR framework so that it does this?

tfwillems · 10-05-2012, 07:21 AM

Thanks to Mark Robinson and Gordon Smyth for answering this

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
DESeq: more than 2 levels per condition?	edue	Bioinformatics	10	09-03-2013 04:52 PM
DESeq high withing group dispersion	wdt	RNA Sequencing	0	07-16-2012 04:54 PM
cuffdiff log2(FC) when 1 condition has 0 counts	crh	Bioinformatics	0	06-25-2012 01:08 PM
DESeq Estimate Dispersion	mattia	Bioinformatics	1	02-09-2012 02:42 AM
condition of use about RNA-Seq normalisation	ickou	Bioinformatics	0	11-23-2011 01:57 AM

10-03-2012, 12:46 PM	#1
tfwillems Junior Member Location: USA Join Date: Jun 2012 Posts: 2	EdgeR condition-specific dispersion I'm dealing with a factorial RNA-seq data set in which cells have been stimulated with various combinations of extra-cellular cues. As such, I was interested in applying the GLM framework in edgeR to assess the contribution of each extra-cellular cue to the differential expression of certain genes. My concern, however, is that both the expression level and the dispersion of each gene varies greatly with the combination of cues. EdgeR doesn't seem to estimate condition-specific dispersion but rather one dispersion per gene (if the tagwise options is used). My question is therefore two-fold: 1) Does it make sense to want to estimate condition-specific dispersions? 2) Is there a way to modify the edgeR framework so that it does this? Last edited by tfwillems; 10-05-2012 at 07:20 AM.

10-05-2012, 07:21 AM	#2
tfwillems Junior Member Location: USA Join Date: Jun 2012 Posts: 2	Thanks to Mark Robinson and Gordon Smyth for answering this