Making a design matrix when the control has a different number of replicates?

Question

[hakimelakhrass]

Hello,

How do you go about making a design matrix when the control has 8 replicates and the rest have 9. Is it possible or do I have to throw out the one extra from the rest. Thank you!

[Carlo Yague]

In most cases you can have 8 replicates vs 9. But what is your goal here ?

[hakimelakhrass]

The goal is to find DEGs between the time points. So a pairwise comparison between all time points.

Answer 1

[andrew.j.skelton73]

Yes you can:

model.matrix(~0 + Treatments, data = data.frame(Treatments = factor(c(rep("Control", times = 8), rep("Treated", times = 9)))))​

[hakimelakhrass]

thanks for that helped!

[1234anjalianjali1234]

if I want to compare cancer v/s normal samples, what is the code for design matrix then? Also, there are different numbers of cancer and normal sample. I have no sample replicates.

[andrew.j.skelton73]

That's a whole different kettle of fish. Depends on your experimental design, if your tumour/normal samples are paired, what type of data you have, etc, etc, etc. The design matrix itself is essentially a description of your experimental design. I think you need to do some more reading, or consult with some bioinformatics support at your institution.

[1234anjalianjali1234]

My samples are not paired. I have no one to consult that's why I am here. If you want then I would like to explain my work.

[andrew.j.skelton73]

Then create a new post if it's a new topic to discuss