Which reference to use for classify.seqs?

I find there are at least three candidate sets. Could you please tell me which one to use and why? Thanks a lot.

  1. trainset14_032015.pds.fasta, trainset14_032015.pds.tax. This is the one used in MiSeq SOP page.
  2. trainset14_032015.rdp.fasta, trainset14_032015.rdp.tax. I do not know the difference between PDS and RDP.
  3. silva.nr_v123.align, silva.nr_v123.tax.

Hi Yangfa,
When I was using mothur for my analysis, I used silva.nr_v123.align, silva.nr_v123.tax. for align.seq command and trainset14_032015.pds.fasta, trainset14_032015.pds.tax(It contains chloroplat, mitochondrial, archae classification also where as RDP and greengenes lack that).
Hope it helps.

BR,
Mashhood

I would encourage you to read the links on this page:
http://www.mothur.org/wiki/Taxonomy_outline

The three taxonomies differ in what they call things. Taxonomy is not an exact science and people have different standards for what they call things. This is reflected in the three general approaches in the rdp, greengenes, and silva taxonomies.

Pat