Nextera XT specific feature classifier

Hi,

in our experiment we're using Nextera XT library for Illumina MiSeq to explore sinonasal microbiota.
We've denoised sequences with DADA2 and I was trying to assign taxonomy to a feature table. I used silva-full-sequence classifier, however results were weird - there were a lot of unclassified sequences.
We've also trimmed sequences to 250 bp, and then the silva-full-sequence-classifier worked like a charm. However, our sequences are around 460 bp length, so we didn't want to lose an additional info.
To tackle the problem of unprecise classification I've decided to train a specific classifier using a RESCRIPt tutorial. I approached a feature extraction step and input a Nextera XT primers:

  • 16S Amplicon PCR Forward Primer = 5'
    TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG
  • 16S Amplicon PCR Reverse Primer = 5'
    GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC

The output of the feature-extraction command was No matches found. What can be the source of the problem? I don't really understand the initial RESCRIPt tutorial database cleaning steps, may that be the issue?

Thank you in advance

Hi @crusher083,

This should help:

-Mike

1 Like

So I mixed adapter sequences and primers per se. After fixing it, everything works like a charm and I got a pretty taxa-barplot. Thank you a lot!

1 Like