I am triying to use MiSeq SOP to analyze my 16S archaeal sequences. It seems to be all good until get to the ‘classify.otu’ step, since when I look into my taxonomic assignments most of them (when not ALL of them) are Bacteria.
I have talked to the sequencing service and we agree that probably our samples are not Archaea abundant and thus the sequencing could have been biased towards bacterial sequences.
However, when looking into the fasta files generated during the analysis with MiSeq SOP process I realize that my sequences are really short after ‘pcr.seqs/screen.seqs’ step.
I was wandering if this could be a problem due to the database I use to align my sequences against, as it is SILVA v132 and it does not contain a lot of archaeal sequences or due to the commands I use, since I use the .oligos file to run this part of the procedure.
Anyone has had a similar problem? Am I doing anything wrong? I have tried to run pcr.seqs against RDP database but it is unaligned and returns some problems with these commands.
I paste all the commands I use and the oligos sequences:
pcr.seqs(fasta=silva.seed_v132.align, oligos=archaea.oligos, keepdots=T)
screen.seqs(fasta=current, count=current, summary=current, start=XXXX, end=YYYY, maxhomop=8)
filter.seqs(fasta=current, vertical=T, trump=.)