extract-reads to train classifier with more than 1 primer pair

I have a couple of questions.

  1. We did full length 16S+18S sequencing. Our data have both 16S and 18S sequences in the same file. I would like to assign taxonomy to this dataset. I am trying to train a classifier for both 16S and 18S combined and would like to know for the step of extracting reads how do I assign more than 1 primer pair?
    Would it look something like this:
    qiime feature-classifier extract-reads
    –i-sequences 16S+18S.qza
    –p-f-primer primerF1 primerF2
    –p-r-primer primerR1 primerR2
    –o-reads 16S+18S_trimmed.qza

  2. On the page: Data resources — QIIME 2 2021.11.0 documentation

Naive Bayes classifiers trained on:

Is this classifier for 16S+18S, or just 16S?

Pre-formatted SILVA reference sequence and taxonomy files:

Are these files for both 16S and 18S or just 16S?

Thank you!

Hi @Gaurav_C,

SILVA has two sets of files, one for SSU and another for LSU. The SSU contains both the 16S & 18S sequences. Fortunately, we've made a great plugin to help you curate your own reference database: RESCRIPt. You work through this SILVA tutorial.

You can either make two different classifiers based on your amplicon region or simply use the premade full-length classifier available on the Data resources page. Check out this post for more details.


An off-topic reply has been split into a new topic: is it possible to train a classifier with multiple genes?

Please keep replies on-topic in the future.

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.