Two genotypes, two ages and three conditions - DESeq2

Question

[annamariabugaj]

I am analysing RNA-Seq data from two different genotypes of adult and pup mice mice collected from three brain regions. I created matrices for each region separately but is it possible to analyse all the data in one run?

My coldata looks like this:

region  id  age genotype 

CA1 N3P81a  pup mut 

CA3 N3P81b  adult   wt 

DG  N3P81c  pup mut 

CA1 N3P82a  adult   mut

CA3 N3P82b  pup mut 

DG  N3P82c  pup wt

etc.

My research questions are:

1. how genotype affects gene expression at each age and each region?
2. which genes are specific for each region in pup mutant?

could I use a design like this: ~ genotype + age + region? or do I need to add interactions? How I should set the contrasts for question nr 1?

I am also thinking about applying the PCATools to explore the data.

[jv]

I am also thinking about applying the PCATools to explore the data.

If you haven't already done QC and exploration, you should absolutely start there before doing differential expression analysis

Answer 1

[swbarnes2]

is it possible to analyse all the data in one run?

I'm not sure what you mean by "one run", but you can throw all your samples together into one deseq object, and then compare the subsets of samples to each other.

I think you need to slow down and break down your research question into what specific questions you want to ask. ~genotype + age + region will likely work, though making columns in colData that combine some of those elements together will make it easier to compare subsets of samples to other subsets of samples. You might need interactions to evaluate other questions, but I'd start by laying out exactly what those questions are.