Here is a link to biostars post of the same question: https://www.biostars.org/p/9558894/#9558894

Haven't received any replies, so thought to try here! Thank you in advance.

We have a prospective study for which at two time points (t1,t2) we collected blood samples and some quantitative traits Q. Using DESeq2, we can ask if the blood gene expression E at t1 or t2 is associated with these traits. In other words, we can test E_t1 ~ Q_t1, then E_t2 ~ Q_t2.

Question: is there a way to test: (E_t2 - E_t1) ~ (Q_t2 - Q_t1) ??? We would like to know if the change in Q is associated with the change in E. Could help me design an equation in DESeq2 for that?

Thank you very much in advance!

We have a time series example in the workflow: rnaseqGene in the Bioc workflows.

But the essential point is the you would put an interaction term between time and whatever trait/condition you have. And then you test that term either with a Wald test (results(dds, name="...")) or LRT (with reduced removing the interaction term).