All archaea after running feature classifier using silva 132 database

[Setiawan (Ruby Setiawan)]

Hello, I have a trouble with feature classifier with silva 132 database in qiime2 2018.8.
First, I followed this tutorial to create classifier from Silva 132 database for 16S gene.
I ran into this thread and found this post to reduce the --p-classify--chunk-size.
I am reducing the --p-classify--chunk-size from 20000 into 10000 but still give MemoryError, then I am reducing the chunk-size into 5000 and the command give classifier.qza results.

Using the results, I am creating the taxonomy file using this plugin qiime feature-classifier classify-sklearn, but the results only shows Archaea.
Can any body give suggestion?

[Nicholas_Bokulich]

Hi @Setiawan ,
This is an issue with the sequences you are using to train the classifier and has been discussed elsewhere on the forum. See this post (and the rest of the thread) for a diagnosis and how to fix:

Wrong 'taxonomy.qzv' file

I think the issue is an aberrantly short sequence being created during the extract-reads step. There is only one taxonomy string that matches the one in question: LAFJ01000960.5.938 D_0__Archaea;D_1__Nanoarchaeaeota;D_2__Nanoarchaeia;D_3__Nanoarchaeales;D_4__Nanoarchaeaceae;D_5__Nanoarchaeum;D_6__hot springs metagenome And here is the 4 nucleotide sequence that is present for it in my extract-reads artifact file (the original full length read doesn’t have any long stretches of N characters, s…

Good luck!

[Setiawan (Ruby Setiawan)]

Thank you, it solves the problem.