Effect of within-group variation on gene expression analysis
1
0
Entering edit mode
hpapoli • 0
@hpapoli-24100
Last seen 16 months ago
Sweden

Hello,

I have RNA-seq data from testis tissue of a natural population of 3 closely related bird species (5 individuals each).

I have detected several thousands (3195) differentially expressed genes in one of my comparisons between the two groups. When looking at some of the differentially expressed genes, I see quite high variation within one of my groups as shown for one example gene below. I have 5 individuals per each group. As can be seen, the group to the right, shows a high level of variation for this gene.

enter image description here

Not all genes show the above pattern. There are many like the one below as well.

enter image description here

My question is that whether in the cases like the first image with large within-group variation, we can really say that the two groups differ and the gene is being differentially expressed since it looks as if there is one individual which has a very high expression value?

Thank you!

DESeq2 • 725 views
ADD COMMENT
0
Entering edit mode

I understand better now. This issue has actually been directly addressed in Deseq2 paper in the section in which the authors talk about shrinkage of log2fold changes in lowly expressed genes (Love et al 2014). When I shrink the log2 fold change for the first case, the log2fold change goes from 6 to 0.23 which would filter out during my analysis. In the second cases, however, the log2fold change does not really change.

ADD REPLY
0
Entering edit mode
@mikelove
Last seen 1 hour ago
United States

Marking as answered...

ADD COMMENT

Login before adding your answer.

Traffic: 752 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