relative abundance > 1% at genus level with only top 5 genera

Hi,
I am trying to have a taxonomy barplot/ bar chart showing only the relative abundance >1% at the genus level with only the top 5 genera.

I have followed the steps shared in various discussions in the form and I am not able to produce that.

First >>> collapse the table
qiime taxa collapse
--i-table ./table_2k.qza
--i-taxonomy ./taxonomy.qza
--o-collapsed-table ./uniform_table.qza
--p-level 6
Note: this input (table_2k.qza) obtained from (qiime feature-table filter-samples)

THEN >>> convert to relative abundance
qiime feature-table relative-frequency \
--i-table ./uniform_table.qza
--o-relative-frequency-table ./relative_frq_6_table.qza
Note: NO filter (p-min-frequency or --p-min-samples)

THEN >>> filter to <1% in at-least 10% of the samples.
Then, I took the collapse table (uniform_table.qza) and I run this:
qiime feature-table filter-features-conditionally
--i-table ./relative_frq_6_table.qza
--p-abundance 0.01
--p-prevalence 0.1
--o-filtered-table ./filtered_relative_frq_6_table.qza

After that, I am stuck, I am not sure how to generate the barplot/ bar-chart showing the relative abundance of >1% at the genus level with only the top 5 genera.

I am not sure if what I did is correct or even makes sense.

Do I need to take the file (filtered_relative_frq_6_table.qza) and generate a new (tazonomy.qza) using qiime feature-classifier classify-sklearn ?

Thank you for all the help!

Hello!
After you found which genera have relative abundances > 1% in 10% of the samples, you can:
Option 1. Create a barplot with uncollapsed and unfiltered table, and click on the color squares in the barplot legend only of features with >1% abundances (that you got by filtering and collapsing). That will show only selected taxa in the barplot with empty spaces for all other taxa.
Option 2. If you want to rescale abundances of selected taxa to 100% (no empty spaces in the barplot), you will need to filter out from uncollapsed table all taxa that have less than 1% abundances and recreate a barplot (I prefer option 1).

Best,

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