Filtering out D_0__Bacteria;__;__;__;__;__ from feature table

User Support
1

Hello,

I am quite new to qiime so was wondering if I could have some help. I have 16S V4 sequences, assigned taxonomy using SILVA 132 at 97%, and have a few samples that are majority D_0__Bacteria;;;;;__ (level 6 taxa bar plot)

image
image508×593 24.2 KB

I was wanting to remove these sequences from my feature table as I found most of them are human after BLASTing them.

However, I followed a forum answer from someone who had a similar issue but found that the command actually retained all those sequences rather than excluding them:

qiime feature-table filter-features
--i-table table-filtered-s26.qza
--m-metadata-file taxonomy-s26.qza
--p-where "Taxon='D_0__Bacteria'"
--o-filtered-table filtered-excluded-table-s26.qza

I realise that the --p-where refers to features to keep and was wondering if there was a way to exclude these instead? I know there is a way to exclude IDs from metadata files but I have over 5000 sequences being classified as D_0__Bacteria.

image
image1292×270 13.9 KB

Thank you

2

Hi @ChloeD93,
See the third example here. qiime taxa filter-table --p-include D_1__ should do exactly what you want! (it will exclude any sequences that are missing phylum-level classification)

Good luck!

3

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