Sizefactors of DESeq2 and edgeR ??? Not equal while using normalization method = RLE of edgeR.

Question

[bioinforupesh2009.au]

Dear all,

I try to compare normalization by edgeR (method=rle) and DESeq2. I assumed it should be equal but its not. Why? Any comment?

## DESeq
# Estimate size factor
dds <- estimateSizeFactors(dds)
sizeFactors<-sizeFactors(dds)

> sizeFactors
       c1        c2             c3             t1               t2            t3
1.0499440 0.8303715 1.2730947 1.0406032 1.1123688 0.8099849

## edgeR (Method of DESeq normalization)
rle <- calcNormFactors(geneCounts.dgelist, method="RLE")
> rle$sample$norm.factors

          c1                c2        c3             t1              t2        t3
1.0398201 1.0503468 1.0296647 0.9079722 0.9871300 0.9921249

Thank you

Answer 1

[elie.maza]

Yes, results are not exactly the same but they are built on the same concept :-) It is clearly explain (I hope) here : http://journal.frontiersin.org/article/10.3389/fgene.2016.00164/full