I am having trouble finding documentation for how to use sortmeRNA with multiple paired-end samples. My goal is to remove rRNA sequences from each of my 8 samples before de novo transcriptome assembly. I see that there is a merge-paired-ends.sh file than many have used, but the sortmeRNA binary that is available on the computer cluster that I have access to (obviously) does not have multiple executable files.
I am torn here. Do I use a different method? I tried mapping all fastq reads to a concatenated LSU/SSU library using bwa (sequences pulled from SILVA) and keeping only unmapped reads but I still had a huge number of over-represented sequences in the fastqc analysis.
If I do go the sortmeRNA route: what is the purpose of merging sequences? My understanding is that the forward and reverse reads are merged, run through sortmerna, produce one output file, and can then be 'unmerged'. Is that correct? Or does merging sequences entail merging all of my samples?
Any insight would be much appreciated. Thank you!