Hi, I have a problem in understanding the correct design formula in DESEq2. I have sequencing coming from different patients with the same disease (Group) but different mutations (Genotype), they are all different individuals (Individuals). My experiment looks like this:

• Group Genotype Individuals
• A WT 1
• A WT 2
• A WT 3
• A WT 4
• A WT 5
• B G1 6
• B G1 7
• B G1 8
• B G2 9
• B G2 10
• B G3 11
• B G3 12
• B G4 13

1 - I want to test for differences between Group A and B by "correct" for genotype information, in other words how can I test for genes that are differentially expressed between condition A and B by correct for Genotype

2 - I also want to test for differentially expressed genes for each Genotype (so for examples DE genes of G1 vs WT)

I have tried with the following design formula:

~ Genotype + Group

But it return me the error "Model matrix not full rank". I think that there is the so called "nested individuals" problems, I have try many different approaches like explained on the DESEq2 manual, but the error still appear. Can you please help me solve this?

Thanks in advantage.

This is directly answered in the vignette section that the error message DESeq2 prints in R points you to read. We have an example of this same design.

The issue is not the same as the nested individuals, because you are not comparing within individuals across groups.