Classify.seqs

Theory behind mothur
1

Hi all,

In the MiSeq SOP step classifying.seqs, if I use Greengenes as the reference database, is the process of classify.seqs:

  1. Was greengenes used as reference when forming the OTUs i.e. closed ref otu picking or
  2. Were de novo otus clustered and then annotated against the greengenes database?

Cheers

Alan

2

Kind of Option 2.

OTUs are constructed de novo in mothur through the (dist.seqs -> cluster -> make.shared) steps. The typical mothur pipeline is to classify all the individual sequences and then assign OTUs a consensus taxonomy of their constituent sequences. If you want to classify the representative OTU sequences themselves, you’d need to do something like the following:

dist.seqs(fasta=XXX.fasta,output=lt,cutoff=0.15)
cluster(phylip=current,cutoff=0.15)
get.oturep(fasta=current,phylip=current,list=current,label=0.03)
classify.seqs(fasta=current)

There’s no direct ability to do closed-reference OTU picking in mothur.