venDiagram after limma showing more genes after specifying up or down regulated specifically
1
0
Entering edit mode
A ▴ 40
@a-14337
Last seen 4 months ago
United Kingdom

Hi all, 

 

was wondering if somebody could explain how limma produces genes that will be entered in to a venndiagram. I have produced a vennDiagram using limma results:

ebayesliversamples<-eBayes(contrastfitliversamples).......

resultsvennliversamples <- decideTests(ebayesliversamples, method="separate", adjust.method="BH", p.value=0.05)

then call vennDiagram to look at the result. This is really nice, however, I get a different number of genes when I then change the "include" argument. 

For "both" up-regulated and down-regulated genes, I get (just as an example) 168 genes on one outer ring (specific to one contrast).

Now, if I ask for genes that are up-regulated only (include="up") then the same outer ring as above now contains 301 genes. Similarly, include= "down" = 174 genes. 

This is really confusing for me, as I thought there should be 168 genes in total and then the up and down regulated genes would be a number out of the 168... 

My question is, why, when specifying to view genes that are specifically up-regulated or down-regulated, are there more genes than the total number of DE genes exclusive to this contrast when specifying "both"

Not sure of the code that might be useful here, I also couldn't directly attach the venn diagrams, however, I will respond with any requests for further code etc! But any help with this would be greatly appreciates as it is incredibly confusing!

Many thanks!

 

venn diagram venndiagram limma microarray differential expression • 1.5k views
ADD COMMENT
1
Entering edit mode
@james-w-macdonald-5106
Last seen 19 hours ago
United States

When you select 'both', the intersection can included genes that go up in one contrast and down in the other. If you select 'up' or 'down', the intersection will now only include those genes that go up or down in both contrasts, and those genes that were up in one contrast and down in the other end up in the unique sections of the diagram.

As an example, say you had 20 genes that were up in one contrast and down in the other. If you select 'both' then those 20 genes will appear in the intersection. If you select 'up' they will now appear in the unique section for the contrast in which they are up-regulated, and if you select 'down' they will appear in the unique section for the other contrast.

ADD COMMENT
0
Entering edit mode

Thank you! In response to your first paragraph, does this then imply that the total number of genes in the venn diagram should be the same? but that genes fall in and out of intersections based on whether you specify up or down? Because when I specify up or down, the total number of genes in the entire venn diagram is completely different to when specifying "both"

Total number of genes for up and down=11966

Total number of genes for specifying both=11605

ADD REPLY
0
Entering edit mode

No, there is no reason to expect the total number to be the same. In the example I gave there are 20 genes that will be counted in both the 'up' and 'down' Venn diagrams (for a total of 40), but that will only be counted once in the 'both' Venn diagram.

ADD REPLY
0
Entering edit mode

Many thanks! and sorry for the delay in replying. Yes this makes sense. Thank you for helping shed light!

ADD REPLY

Login before adding your answer.

Traffic: 475 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6