Hello Biostars, I tried to look elsewhere but I am at a loss I hope someone here can help me. I was given Affymetrix Clariom expression data for a differential expression analysis for four different groups:

1. Infected-plasmid
2. Infected-empty
3. NotInfected-plasmid
4. NotInfected-empty

After Robust multi-array RMA normalization using Affymetrix software. The person who gave me the data asked me to to a fold change difference of this category:

• (NotInfected-plasmid x Infected-plasmid (fold)) X (NotInfected-empty x Infected-plasmid (fold)) (fold difference)

Is there such a thing as a "two step differential expression" or am I just crazy and I can't do a fold of a fold difference and I should only do a standard case-control regrouping my categories two by two?

So far I have tried the gage package with gagePipe gageComp function which gives me the common significant pathways between many differential expression analysis.

If someone could help me I would be grateful.

Thank you François