I'm trying to find a way to examine the similarity of sequential samples taken from a patient. Let's say, that I have 10 patients (A, B, C, D etc.) and from each I have three samples taken at different time points (1, 2, 3). I have gene expression data from each sample. Now I would like to find out whether the three samples from each patient are similar. I have done clustering, differential gene expression analysis (paired) and correlation analyses, and I can see that the samples correlate, but there are also differences. Is there a way to determine whether samples 1-3 from a patient are statistically similar to each other? So that similar assumptions can be made based on, for instance, sample 3 (patient A) as from sample 1 (patient A).
Isn't it sufficient to take all DEGs, put into hclust and see that samples per patient are closer to each other than to the rest? You don't have to put a pvalue on everything, clustering is perfectly valid to show similarities if you ask me.