over 70 % of reads are assigned as d__Bacteria;__;__;__;__;__;__ qiime feature-classifier classify-sklearn

Hi @jrhaulung,

Two things,

It could be that the premade curated database does not contain the V1V2 sequences, as they've been excluded during strict curation, due to being at the end of the sequence. Which means many reference sequences may have been removed. You can make your own database as outlined here.

Otherwise I suspect a read orientation issue... if you search the forum, you'll see many solutions to this problem. I'd suggest starting with this approach.

3 Likes