Visualize overlap differentially expressed genes of multiple comparisons.

Question

[lucatucciarone]

Hello everyone,

I need to represent the overlap of differentially expressed genes across multiple comparisons.

Essentially, I have the log2foldchange of my 4 samples (Control, Treatment 1, 2, 3). Obviously my comparisons are Control vs Treatment 1 Control vs Treatment 2 Control vs Treatment 3 Control vs Treatment 4

Now I want to understand how those 3 comparisons overlap too each other. How many up-regulated genes are commonly up-regulated by every treatment? How many just in the 1st treatment? and so forth.

Now I can see that I can create a sort of monster of Venn diagram by have 8 different "clouds" to intersect (4 comparisons up-regulated and down-regulated).

Is there a better way to represent my data?

[ATpoint]

You should check the keyword upset plot.

[lucatucciarone]

mmmh that's a good idea, I've already used them.

There is nothing specific for up and down regulated genes though?

[russhh]

I tend to extract the collection of significant genes (either defined by F test or by union over the comparisons), plot the log-fold-changes wrt each contrast on a heatmap and indicate singificance for a given contrast with a black/white row annotation (one row annotation for each contrast). Even with just two contrasts, distinguishing up- and down-regulated genes can lead to overly complicated venn diagrams.

[lucatucciarone]

This seems more doable. Would you mind uploading a sample picture to see how it turn out?

Answer 1

[Liang Sun]

you can use this online tool DiVenn which will generate a nice graph https://divenn.noble.org