Can heatmaps be compared across multiple different experiments? Obviously, you can compare sample runs within a single heatmap, but what about comparing sample runs in one heatmap to sample runs in another heatmap (e.g., if I want to track the activity of a gene across multiple therapeutic treatments which I performed in separate experiments from last year)? I am worried that I may run into statistical issues when doing this kind of cross-comparison, since as far as I'm aware the heatmap and heatmap.2 commands in R perform the proper coloring of the heatmaps within a single dataset (not multiple datasets).
I would merge the data into the same heatmap. If you have a sample that was included in both the RNA-seq from a year ago and the more recent run you should compare those first to see what the batch to batch variation is like. If they are very similar (R^2 > 0.9) then maybe you can just go ahead just try merging all the samples in the same heatmap. If there's significant variation or you don't have a common replicate you may try normalizing or scaling all the data to fall within the same range of values then generating the heatmap.