Dear all I have bulk RNAseq data and my goal is to detect genes that change over time after a treatment.I have the following design
I have 7 subjects and 5 time points. One time point is baseline (aka pre treatment), but I actually only have one treatment arm. Given the fact that I have subjects not being measured at some timepoints and that I want to account for intraclass correlation within subjects I go with mixed models. I would also like to use limma to take advantage of voom and ebayes
My question here is the following: .
In a linear mixed model like with lmer I would adjust for baseline eg. to increase power . I know that limma is not exactly like lmer since it uses covariance matrix (like compound symmetry I presume, but my question is.. Is there a way to adjust for baseline gene expression in my case?
I had a look at that post . But didnt help me, since there are more like one treatment and different questions