DESeq2 design matrix help

[Chandu]

Hello All,
I am a bit confused and lost in DESeq2 design matrix in my experiment.

Here is the description of my experiment.

Total samples = 12
Number of mice = 6 ( two RNA-seq from each mouse, one from head and another from body)
Number of genotypes = 2 ( Wt_a and a_b)

I wanted to make two following comparisons. Each group has three replicates. We know that the same genotype on head and body have different effect.
1) Wt_a_body Vs Wt_a_head
2) a_b_body Vs a_b_head

These comparisons gave too many differentially expressed genes, I think this is because of different organs (body and head).

What is the appropriate design matrix, for removing the body vs head differences, while comparing the genotype difference between the organs?

I listed below, all the sample and sample groups with replicates.

SampleName SampleGroup SampleSource Replicate
E1_body Wt_a_body body 1
E2_body a_b_body body 1
E3_body Wt_a_body body 2
E4_body a_b_body body 2
E5_body Wt_a_body body 3
E6_body a_b_body body 3
E1_head Wt_a_head head 1
E2_head a_b_head head 1
E3_head Wt_a_head head 2
E4_head a_b_head head 2
E5_head Wt_a_head head 3
E6_head a_b_head head 3

Many thanks in advance.

[dpryan]

~replicate+SampleSource+SampleGroup

Be sure that "replicate" is a factor. This assumes that there's no interaction between the source and group, though you could include that as well.

[Chandu]

Thank you Ryan.