Condition-specific genes

[SylvainL]

Hi all,

I have to analyze RNAseq data concerning 4 different tissues. I have biological replicates.
Until now, I did pairwise DE analyses using edgeR and I got the Differntially Expressed genes for each comparison, with p-values and FDR.

Now, I would like to look at the condition-specific genes but I am not sure about the way to get them...
I can either start from the different fold-changes in the different pairwise comparisons and try to get the conditions were a gene is the most expressed, but it is quite complicated.
I also tried to cluster the genes (after sequencing depths normalizations) but then I have the problem of genes which are not expressed at all (0 counts) in some tissue.

I looked at some R packages like SpeCond but this one was done for microarrays so the quantification distribution is quite different and I do not think we can apply the same statistical model.

How do you do?

Thanks in advance...

[Gordon Smyth]

It is best to build up the condition-specific genes by making comparisons between the conditions.

Suppose you want genes that are specific to condition A. For a very stringent set, choose genes that are DE and up-regulated for all the comparisons A vs B, A vs C and A vs D.

For a less stringent set, find genes that are DE for the contrast A - (B+C+D)/3.