I wanted to get your advice on using Deseq for DE for RNA SEQ results for several strains of the gram negative anaerobe B. fragilis.  We have three pathogenic strains and two lab strains, similar but not identical genomes.  This is not yet a very popular use for comparison transcriptome analysis but we actually have found it useful and highlighted several genes that may be important in virulence.

My question:   Is it reasonable to eliminate the rna counts for the 5s, 16 s 23 s and tRNA from the data submitted to deseq; I am interested in comparing CDS; either transcripts/genes only present in pathogens or transcripts/genes present in both pathogens and lab strains but up or down regulated significantly in one group.

Thank you in advance for any advice you can give me.

It is certainly reasonable to disregard counts for rRNA and tRNA if you are only interested in coding mRNA. Most likely, however, it will make little difference whether you remove the rows concerning these from the count matrix before running DESeq2 or simply ignore the p values you get for them in the result returned by DESeq2.