rna-seq/DEseq2 analysis - is it a linear combination problem?

Question

[aabbcc1]

Hi, I need some advice concerning my data analysis. It's RNA-seq data looking at gene expression differences with a treatment. Also recorded is the sampling time in hours for each sample that I would like to include as a covariate.

I have an experimental design which looks like the following:

treatment sample_time

control 13

control 14

control 15

control 15

control 15

control 16

control 16

control 17

treated 19

treated 19

treated 19

treated 19

treated 20

treated 21

treated 21

treated 22

I thought about doing ~ sample_time + treatment but I am thinking that I might have problem with all treated samples collected later than the control ones and being unable to separate that from the effects of the treatment. What would be the best way to approach this?

Thanks

Answer 1

[ATpoint]

I thought about doing ~ sample_time + treatment but I am thinking that I might have problem with all treated samples collected later than the control ones and being unable to separate that from the effects of the treatment. What would be the best way to approach this?

Yes, you cannot control for sample time here as they are different in treated and control, so it comes down to ~treatment for your model.