Question: Deseq2 Interaction Term to catch
gravatar for guipagui
22 days ago by
guipagui0 wrote:


I read the documentation, vignette, tuto ... but i'm face an issue.

Let's try :

dds <- makeExampleDESeqDataSet(n=10000,m=12)
dds$condition <- factor( c( rep("Ctrl",6), rep("Trt",6) ) )
dds$genotype <- factor(rep(rep(c("WT","MU"),each=3),2))

dds$genotype = relevel( dds$genotype, "WT")

If i want to modelise the effect of treatment for genotypes:

Méthode A :

design(dds) <- ~ genotype + condition + genotype:condition

res = results(dds, name="genotypeMU.conditionTrt")

that's ok but :

Méthode B

dds$group <- factor(paste0(dds$genotype, dds$condition))
design(dds) <- ~ group

res = results(dds, name="group_WTTrt_vs_MUCtrl") doesn't give the same result

What's i'm wrong ?


ADD COMMENTlink modified 21 days ago • written 22 days ago by guipagui0
gravatar for Michael Love
22 days ago by
Michael Love19k
United States
Michael Love19k wrote:

Please look at the diagram in the vignette in the section on interactions.

The most recent DESeq2 question on the support site had the exact same question.

Above you are mixing up main effects and interactions. The interaction is not a main effect, but a difference between the two condition effects.

ADD COMMENTlink written 22 days ago by Michael Love19k

Thanks a lot,

The test :

res = results(dds, name="genotypeMU.conditionTrt")

answer to question "Effect of treatment for the 2 genotypes" ? is right ?

I don't see how make it with methode B ... ? it's possible ?

ADD REPLYlink modified 21 days ago • written 21 days ago by guipagui0

No it’s much more difficult to test the interaction with method B.

ADD REPLYlink modified 21 days ago • written 21 days ago by Michael Love19k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 198 users visited in the last hour