This is a test version of Biostars. For the public version, visit https://www.biostars.org.
contrast matrix design for a continuous variable at multiple groups

Hi all,

I am trying to figure out signals from DNA methylation for a continuous variable at multiple groups. We have mice samples at 4 time points in both sexes, and we are testing if DNA methylation change with V1 using limma

I used the following design matrix and added duplicateCorrelation for random effect.

> X = ns(time_point, df = 3) 

> design = model.matrix(~ V1*sex + X)

I am asking the questions:

  1. which DNA methylation positions respond to V1 in females?
  2. which DNA methylation positions respond to V1 in males?
  3. which DNA methylation positions respond to V1 differently between females and males?

Based on the questions, does this design make any sense?

design contrast limma
• 1,315 views • link
updated by Asaf Asaf • written by dtzhu337 • dtzhu337

1 answer

The element V1 will give you the overall effect of V1. V1:Sex interaction will answer the question how V1 is different between males and females. If V1 has an effect you can then see how V1:sex is behaving on that gene and conclude how it's affecting males and females differently.

• 0 views • link
written by Asaf Asaf

Hi Asaf,

Thank you very much. One step further, actually I am not interested how DNA methylation levels respond to age, can I simplify the model and ignore age splines as:

design = model.matrix(~ sex + sex:V1)

Thank you

• 0 views • link
written by dtzhu337 • dtzhu337

It doesn't matter if you care about this information but rather if the age has an influence. If it does then you will want to put this in your model.

• 0 views • link
written by Asaf Asaf

Log in to answer this question.

More posts like this