Hi,

I am performing a RNA-seq study on embryonal cell. All samples divide into two sets: KO and WT groups. Within each group, each control sample is paired with another sample which have been subjected to treatment, the two samples derived from the same patient. Below, I am reporting an example of the samples

I would like to perform a paired analysis to make comparison among the 4 groups ( KO.Treatment , KO.CTRL, WT.Treatment and WT.CTRL. The problem is that Patient cannot be considered in the model design as it is linear with Condition variable. Should I add a Nested variable (as in the following table) in the design? ~ Condition + Condition:Nested would be the right design formula in this case?

Thank you in advance.

Giulia

See section 3.5 of the edgeR User's Guide.

Thank you James!

Hence, from what I have understood, I have to split the variable Condition into Genotype and Treatment variables and I have to include them into the formula together with Nested variable. Then to understand the effect of treatment intra-genotype, I have to design a formula ~ Genotype +Genotype:Nested + Genotype:Treatment?

Best,

Giulia