How to make a circular heatmap tree

[macrobiome (T)]

I am running QIIME2-2023.2 in a conda environment.

I have two questions:

1. Is it possible to get a heatmap tree (such as the one in part d of the below figure) through beta diversity analyses?

2. How do I combine data from multiple samples to get a heatmap tree like above? Let's say I have sequencing data from 7 samples and I want to combine them to generate one figure like above.

Thank you!

[crusher083 (Valentyn Bezshapkin)]

Hello,

Moving your topic to general discussion, there are no way to do it in Q2, possibilities to replicate this visual include graPhlAn or iTOL.

Cheers
V

[macrobiome (T)]

Thank you! Is it possible to generate a singular phylogenetic output obtained from multiple samples through Q2 which I can subsequently plug in to the tools you mentioned? Thank you!

[crusher083 (Valentyn Bezshapkin)]

Yes, the tree can be generated from Q2. I am not sure what data are represented on the heatmap, but probably possible to extract it as well.

[SoilRotifer (Mike Robeson)]

Hi @macrobiome, you should be able to make a figure similar to e based on the information
here.

Then you can make d using q2-empress. We have a YouTube video, that provides an example, specifically at time point 9:20.

-Mike

[macrobiome (T)]

Hi @SoilRotifer, thanks for your note. I ran the following code to generate a heatmap based on the page you linked (This was the last command; I followed the steps prior to this exactly)

qiime feature-table filter-features /
--i-table heartonly-merged-table.qza /
--m-metadata-file heartonly-transposed-table.qza /
--o-filtered-table heartonly-filtered-table.qza'

I got the following output without beta diversity (My apologies if this a dumb question)

heatmap.qzv (2.1 MB)

What did I do wrong?

[SoilRotifer (Mike Robeson)]

@macrobiome

I got the following output

Nice work!

@macrobiome

without beta diversity

You won't be able to make a heat map with beta-diversity values in this way. Remember beta diversity is a pair-wise metric. Meaning, that beta-diversity it is the distance between any two samples. Thus, you cannot have a per-sample value, which is what alpha-diversity can provide.

[macrobiome (T)]

Would you be able to suggest on how to make a heatmap with beta-diversity values similar to figure e in the earlier post?

Secondly, I see a bunch of nodes in the qzv (image below). What do they represent? I haven't seen them in any papers.

[crusher083 (Valentyn Bezshapkin)]

Re1: use a seaborn library, function clustermap seaborn.clustermap — seaborn 0.13.2 documentation
Re2: Those are hashes, which are used as a compact representation of original ASV sequences (writing 150 letters of original ASV sequence is not optimal)

Cheers
V