Hello,

I am new to DE analysis, and I also looked for the answer to my question prior to writing, so sorry if there is a similar thread already and I couldn't find it.

I am trying to analyze the data from the RNA-seq analysis of the 5 different groups (A, B, C, D and E) with 3 biological replicates each. I would like to compare all my groups (B,C,D,E) to my control (A). My question now is: Is it better to compare the 2 specific groups at the time (B vs A, C vs A...) using the input raw counts file containing all the groups and replicates (e.g. A1, A2, A2, B1...) using the DESeq2 result function like this: res<- results(dds, name = "conditionBvs_A")?

Or, is it better to first create the file containing only the raw counts of A and B, and then perform DESeq analysis, using that input file only, and without specifying any 'name' in the result function, as following? res<- results(dds)

I tried both, and I obtained different numbers of DE genes.

I now don't understand why there is this difference (due to the different normalization possibly?), and which method is better to use?

Thank you, any help is appreciated. Snezana

This is one of the frequently asked questions in the vignette (see FAQ at the end).