Question: vennDiagram in Limma package
0
12 weeks ago by
MicheK0
MicheK0 wrote:

Hello,

I have a question regarding the vennDiagram function in the limma package.

I'm trying to make a Venn Diagram comparing 3 groups relative to a control (4th group). I.e. group 1 vs group 4, group 2 vs group 4, group 3 vs group 4. My code is the following:

fit_contrast_venn <- contrasts.fit(fit_filter, contrast.matrix_v2)
efit_venn <- eBayes(fit_contrast_venn)
vennDiagram(venn_results, "include"="up")
vennDiagram(venn_results, "include"="down")


When I make the venn diagrams for up-regulated genes, am I making a Venn diagram for only the genes that are upregulated for all 3 groups (group 1, 2, and 3)? What happens to genes that are upregulated in group 1 but downregulated in group 2 or 3? WIll these also show up on the venn diagram?

I'm just trying to understand the Venn Diagram that I am producing.

Thank you,

limma • 123 views
modified 12 weeks ago by James W. MacDonald49k • written 12 weeks ago by MicheK0
2
12 weeks ago by
United States
James W. MacDonald49k wrote:

When you use the include argument, you can interpret 'up' and 'down' to mean 'show me only those genes that are up or down-regulated'. So if you have a gene that is up in group 1 and down in groups 2 and 3, that gene will only appear in the unique section of the group 1 portion of the Venn diagram. if you say 'down', it will be in the intersection between groups 2 and 3, and if you say 'both' it will be in the central intersection between all groups.

Thank you.

Similarly, if a gene is upregulated in group 1 but not changed in groups 2 and 3, will it show up in the unique section for group 1? Also, does that mean the same gene can show up in both upregulated venn diagram and downregulated venn diagram?

It's a logic thing that you should be able to figure out for yourself. If you have three comparisons, then for each gene you have three choices; up-regulated, down-regulated, not changed. If you choose 'up', then you will ignore everything but the up-regulated comparisons. If you choose 'down', you ignore all but the down-regulated.

So if you have a gene that is up in one comparison and down in the other two, what happens when you choose 'up'? What about 'down'?