Hey all,

I have the following problem, maybe you can give me some hints ....

I have 2 batches with one batch 1 sample per individual and the other batch 3 samples per individual.

I have combined both batches in one model and I am adjusting by batch ~ batch + Gender.

Now I wonder whether I need to do anything about the second batch since each individual has 3 samples in there...

How would you treat this problem? And is there any setting in DESEQ2 to take care of that?

Thank you, Bine

We can model individual with fixed effects, but it will give more weight to the individual sampled three times where you probably want inference to be across individuals, not samples. This is like a random effect in mixed effects modeling. See duplicateCorrelation in limma package.