I am trying to implement a pipeline for RNA-seq Analysis using the Tuxedo suite of tools. I am planning on doing the following Cufflinks ---> CuffMerge ---> CuffDiff. The analysis I am trying to conduct contains 2 different conditions with two samples each for a total of four files.
Condition 1: Sample 1 & 2 Condition 2: Sample 3 & 4
My question is for the CuffMerge step should I be merging all four samples or just do merges within the conditions?
All 4 of these samples are from the same cell line so I assume that the best practice would be to merge all 4 samples and then use CuffDiff to calculate the FPKM values at the condition and sample level. Is this a correct assumption?