Hi Dear 24/7 Admins,
As I understood a discussion in forum, you had not fixed the Heat Map size in Feb 2018. I am having a problem concerning the heat map size. It is as long as the Nile river indeed. Is it still remained unfixed? What is your suggestion to overcome the issue?
Sincerely yours,
Filter out low-abundance features using feature-table filter-features:
https://docs.qiime2.org/2019.1/tutorials/filtering/#total-frequency-based-filtering
Does the filtration remove unassigned taxa as well?
The commands employed for this purpose:
qiime feature-table filter-features
–i-table TableDenoisedLibA.qza
–p-min-frequency 2
–o-filtered-table filtered-table.qza
qiime feature-table heatmap
–i-table filtered-table.qza
–m-metadata-file Metadataforw1and4.tsv
–m-metadata-column SampleType
–o-visualization filteredheatmaplibA
–verbose
but I got the previous result. No change!
The feature statistics:
Frequency per feature
Frequency
Minimum frequency 2.0
1st quartile 16.0
Median frequency 37.0
3rd quartile 100.25
Maximum frequency 13,437.0
Mean frequency 159.43719008264463
I thought it would be helpful in giving a feedback!
Thanks
Also, I need to know this command *qiime feature-table filter-samples * proceeds with this command *qiime feature-table filter-features * in the tutorial/ feature-table plugin. Should I use sample filtering then feature filtering or it is not recommended?
Hey!
I think you are making progress and I think you are very close to answering your own question. I'm going to give you two hints, then let you keep working on it.
Frequency per feature
Frequency
Minimum frequency 2.0
–p-min-frequency 2
but I got the previous result. No change!
I think I know why!
Should I use sample filtering then feature filtering or it is not recommended?
You would use either one, or both, in any order. All options work.
Also, you might like this plugin:
https://docs.qiime2.org/2019.1/plugins/available/feature-table/merge-taxa/
Keep up the good work,
Colin
I garbed the point! 
So I am wondering what number I should add! 2 is lowest number! but I must increase it up to what? Is the value considered arbitrary? Or I have to pick it up logically?
Thanks
Dear @colinbrislawn
This time is used:
qiime feature-table filter-features
–i-table TableDenoisedLibA.qza
*–p-min-frequency 37 *
–o-filtered-table filteredtable37
qiime feature-table heatmap
–i-table filteredtable37.qza
–m-metadata-file Metadataforw1and4.tsv
–m-metadata-column SampleType
–o-visualization filteredheatmaplibA37
–verbose
But it is still long! it did fix!
Whoops, I gave you the wrong link. Check this out:
qiime taxa collapse -h
I would try something like this:
qiime taxa collapse
qiime feature-table filter-features
qiime feature-table heatmap
Then I would try other values of --p-min-frequency and other levels for qiime taxa collapse.
Colin
Hi,
These command were used:
qiime taxa collapse
--i-table TableDenoisedLibA.qza
--i-taxonomy taxonomylibA.qza
--p-level 1
--o-collapsed-table collapsedtableking.qza
--verbose
qiime feature-table filter-features
--i-table TableDenoisedLibA.qza
--p-min-frequency 37000
--o-filtered-table feturefiltered37000.qza
qiime feature-table heatmap
--i-table feturefiltered37000.qza
--m-metadata-file Metadataforw1and4.tsv
--m-metadata-column SampleType
--o-visualization heatmaplibA37000
--verbose
And this is the result:
Hi Meha,
You are selecting from your ‘TableDenoisedLibA.qza’ any ASVs with a total count larger (or equal ) than 37,000. Have you visualised how many ASVs are included into the ‘feturefiltered37000.qza’?
Also, sorry for my silly question, I’m not sure why you creating the ‘collapsedtableking.qza’ and not using it further? To me, is not a bad plan to collapse the data at level 3 (or maybe 4) and use this collapsed data for the heatmap creation.
Luca
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.
