[BioC] DESeq and EdgeR log fold differences

[Steve Lianoglou]

Hi Ioanni,

Quick thought:

On Thu, Nov 29, 2012 at 5:06 PM, Ioannis Vlachos <iv at on.gr> wrote:
> Hello everyone,
>
> I thought of conducting a parallel DE analysis with EdgeR and DESeq using a
> dataset that I have been working on lately.
>
> The dataset has two conditions with two biological replicates each.
>
> Let's say: Wild Type, Wild Type, Knock Out, Knock Out.
>
> It's a smallRNA-Seq dataset, mapped to miRNAs.
>
> I have tried various analyses using both programs and I have noticed this.
>
> There are very large differences in fold changes for some miRNAs between the
> two programs, even when using "RLE" for EdgeR normalization.
[snip]

Have you tried doing a scatterplot of the DESeq logFC vs the edgeR
logFC for the same genes?

Is the range systematically compressed in one comparison vs. the other, or?
Maybe its a function of the `baseMean` values?

-steve

-- 
Steve Lianoglou
Graduate Student: Computational Systems Biology
 | Memorial Sloan-Kettering Cancer Center
 | Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact