I want to check the differential expression of a specific class of transcripts (say, long non-coding RNAs) using DESeq2. Now, I know that the normalization step takes into account the total number of reads per sample. If I only input a subset of all the transcripts, the sequencing depth will change.
So my question is: can I simply input the subset of interest? Or should I input all transcripts, and subset the results later?