Accounting for fish body mass in DE?
0
Entering edit mode
@e2642572
Last seen 6 weeks ago

Hi,

I'm currently using EdgeR to look at DE in fishes in response to a toxicant and a control. I'm using a pairwise approach looking at control-treatment comparisons within 3 time points (e.g. C vs T @ 12 h, C vs T @ 24 h, etc). I'm noticing some potential effects of body mass on driving some of the relationships in driving the gene responses and wanted to include each individual's mass as a factor in the model. However, I've been trying to find documentation on the proper way to set this up in the model and have been unable to locate anything. I was wondering if anyone could point me in the right direction for previous examples. I was thinking it may be a similar setup to the batch effect example provided in the user's guide but am unsure.

Edit: I currently have my model matrix setup as the following testing for treatment/time effects:


design <- model.matrix(~0 + Nic_Gill_edgeR$samples$group)

GeneExpression edgeR • 143 views
1
Entering edit mode
@james-w-macdonald-5106
Last seen 2 hours ago
United States

Not knowing what is in Nic_Gill_edgeR$samples$group, that particular detail isn't particularly helpful.

Anyway, assuming that your group is basically a combination of time and treatment, and you are using a contrast matrix to make comparisons, you can simply add another column to your design matrix containing the body mass of the fish. You will then need to add a row of zeros to your contrast matrix (or simply use makeContrasts, which will handle the details for you).