I am trying to use deseq2 to compare differential gene expression across multiple generations of repeated treatment. I have three replicates for each condition and there is a treated and untreated/control condition for each sample. In total these are my conditions:

• 1 time treatment: control vs treated
• 2 time treatment: control vs treated
• 3 time treatment: control vs treated

So far, I have been running deseq2 on each generation separately since each generation has their own control. However, I want to be able to compare between generations and was wondering if it was possible to do so/run deseq2 on them at the same time? Hypothetically, all the controls across generations should be the same because they are all untreated WT cells. Is there a way to normalize these datasets across multiple generations so that I can compare them together?

Thank you!

Assuming that all experiments are performed the same but at different times you can create a factorial column timeOfExperiment and simply include that into the design. That will account for the differences due to time (so unwanted variation) while testing the factors of interest. See the vignettes on batch effects. That is the same principle here. In the end you would run ~timeofExperiment + (...) where ... is the design you want to test.