Hello,
I have some RNA-seq datasets of similar experiments from different studies, no replicates in between studies, and not the same sampling points. In this way, there are 5 studies with 3, 2, 4, 5, and 2 treatments/samples each, with no sample replicates. I've pre-processed all of them the same and mapped them against the same reference genome. So now I've got the reads count matrix but their scale differs between studies. My thought is that I have to somehow normalize them proportionally and the way I see to do that is based on housekeeping genes. There're some reference housekeeping genes for this organism, I've checked their counts manually and despite the differences between studies, their counts are coherent to housekeeping genes.
That's where I am standing. I've read this tutorial and I think it's something on this way. In my case, I can't categorize samples just as treated/untreated as there are more than 2 treatments.
I recognize this question is too vast and I don't expect an exact answer, but I would like some guidance on which way to go because unfortunately, I'm not getting any more advances in my researches.
I wonder if this can be called metatranscriptomics if you are aligning against a single reference.
What exactly do you hope to get out of this analysis?
I've done my own experiment and I would like to reinforce the behavior of my genes of interest using other data with similar experimental conditions
I get your first point now. It's not metatranscriptomics indeed, it's a transcriptome meta-analysis.
Perhaps searching for the right key-words will help me better. My mistake.