Hi!

So I'm basically asking the same question as this post, but have a slightly more complex design and want to make sure I'm setting up the proper contrasts.

I have samples from two different genotypes (B6 & TCR) reared in 3 stress conditions (no stress, NS; single hit, SH; double hit, DH). I am interested in the main effect of genotype, the main effect of stress, and the interaction between the two and so have coded my design like so: ~ genotype+ DoubleHit+ Genotype:DoubleHit.

I understand that retrieving results as follows:

res = results(dds, name = 'GENOTYPE_TCR_vs_B6')

only gives me the effect of genotype within the reference level of stress. I have already releveled the factor in the metadata such that no stress is the reference level.

I want the main effect of genotype independent of stress and vice versa.

I have resultsNames as follows:

[1] "Intercept"               "GENOTYPE_TCR_vs_B6"      "DoubleHit_SH_vs_NS"     
[4] "DoubleHit_DH_vs_NS"      "GENOTYPETCR.DoubleHitSH" "GENOTYPETCR.DoubleHitDH"

What is the correct way to set up contrasts in the results function to get the desired outcome?

My attempt at getting the main effect of genotype independent of stress:

res = results(dds, contrast=c(0,1,0,0,0.5,0.5))

And vice versa:

res = results(dds, contrast=c(0,0,1,1,0.5,0.5))

If I'm on track or way off base please let me know! Furthermore, how would setting the contrast to 1 for only one stress condition (ex. if I'm only interested in DH vs NS) compare to setting them both to one for interpretation?

Thanks so much in advance!

Unfortunately, I have to restrict my time on the support site to software related questions. I recommend for statistical consultation about analysis and interpretation of results, working with a local statistician or someone familiar with linear models in R.