How to create a Venn diagram from core-feature and from biom table?

[Antani (Edoardo Scali)]

Hi everyone,

I am trying to create a Venn diagram that shows the number of taxa shared between my groups: A, B, and C.

I tried to follow suggestions from a similar discussion to address this problem. I attempted to create a core-feature table for the feature-table that includes all A, B, and C groups, as well as for a feature-table obtained by filtering each group. However, I obtained a table with all zeros for the core-features, which included feature counts = 0, fraction of features = 0.0, and feature list = "No core features" for all the steps.

Do you have any explanation for why I am getting these results?
Alternatively, do you know if I can obtain a Venn diagram from a biom table?

Thank you so much for your help.

[colinbrislawn (Colin J Brislawn)]

Hello @Antani,

@Antani

However, I obtained a table with all zeros for the core-features, which included feature counts = 0, fraction of features = 0.0, and feature list = "No core features" for all the steps.

Well, that's no good! Can you post the full command you ran so we can see what may have caused this?

P.S. Venn diagrams have some issues. Instead of using a Venn diagram, you could try using other methods to compare the taxa between your three groups, like the Emperor plots made by qiime diversity core-metrics-phylogenetic (example from the PD-mice tutorial).

[devonorourke]

Not sure if this will make the matter better or worse, but in cases where my goal is to show shared or distinct characters among sets/groups, I tend to resort to a style of plot shown here. Hope this doesn't upset @colinbrislawn 's important point that existing alternatives may be more appropriate than my proposed visualization

[Antani (Edoardo Scali)]

Hello,
thank you so much for your help.
First I've create three different table.qza, one per each of the three groups I have.

I've used the following code:

qiime feature-table filter-sample \
  --i-table table.qza \
  --m-metadata-file sample-metadata.tsv \ 
  --p-where "[group]='A'" \
  --o-filtered-table table-A.qza

here is the code I've used:

qiime feature-table core-features \
  --i-table table-A.qza \
  --o-visualization core-feature.qzv 

I've repeted filtering and core-features for all the three groups, and I've obtained a table with all zeros for the core-features, which included feature counts = 0, fraction of features = 0.0, and feature list = "No core features" for all the three filtered tables.

Basically, what I am searching is a way of represeinting how many microbs are shared amongs the groups I have after the classification. I thought that a Venn diagram may be a good option.
Do you have any reccomendation?

@devonorourke thank you for your suggestion, I'll check the resurce you shared.