Giant brainstorm thread - your ideas for microbial ecology data visualizations!

[hollybik]

What types of data visualizations would be useful for the analysis or interpretation of your microbial ecology or metabarcoding datasets? We want ideas for data visualizations that would be useful for researchers, educators, citizen scientists, and any other type of end users (e.g. technical to lay audiences).

Brainstorm your ideas below (and feel free to include example figures/images/etc. to help illustrate your point). Let us know what frustrates you with visualizations in existing tools (QIIME, phloseq, R packages, etc.) and tell us about data viz tools/ideas that don't currently exist.

[manasvil]

I was at a Data Vis meetup in NY and they were talking about unusual data visualizations in the lift sciences and here were some that were interesting:

Target Plot

Volcano plot

Ellipse Plot

I am going to try and find some examples with explanations so they're more useful. But for now leaving this here.

[manasvil]

@hollybik I ended up writing a post about Target plots: https://medium.com/axis-group-visual-analytics-practice/that-which-we-call-a-target-plot-e295587009a9 in case you ever need more detail on what they are. Funny enough, I could not find much material on them during my search.

[tomarashish]

Hi @hollybik

I created a prototype web app for metagenomics data comparison using sunburst. The Web app shows the reference taxonomy as a tree view and multiple samples taxonomy hierarchy as sunburst. Following is a screenshot of the web app. Will like to know your view?

[tomarashish]

Hi @hollybik
Will like to hear from you about the visualization mentioned above.

[hollybik]

Hi @tomarashish, thanks for the examples - in the original Phinch framework we did have a tree view type visualization (called Sankey diagrams). The sunburst idea is also interesting. I'll pass on this idea to our developers to see if we might be able to incorporate something similar into the new Phinch desktop app.

[geboro]

The donut-partition visualization that was available in Phinch v1 was very clear for exploratory comparative analyses. Also, the bubble charts were very useful when talking to non-academic audiences.

[hollybik]

Thanks for the above feedback! We'll keep this in mind as we move forward with new visualization features for the Phinch desktop app.