I was wondering if someone could explain what the "linear modelling" part in Limma (linear models for microarray data) does. I've done a lot of reading in manuals already and nowhere it seems to make me understand what is done and why.

The manual says this:

You have to start by fitting a linear model to your data which fully models the "systematic part" of your data

And I learned that a linear model is constructed for each gene

The question that I have is, what does this model look like and what is 'plotted' in this model, for example what is on the X- and Y-axes? I can't seem to find it anywhere.

Thanks in advance,

Christian

This is a site for software questions rather than statistical education. If you want to learn about linear models, there are resources available elsewhere - a quick Google of "linear models" gives a number of helpful links. Here's one I prepared earlier.

what does this model look like

It looks like the first equation here.

what is 'plotted' in this model, for example what is on the X- and Y-axes?

There is no general association between a linear model and a specific visualization (except in the simplest case of simple linear regression). There are multiple explanatory variables you could put on the x-axis, and even the y-axis may not necessarily show the response variable, e.g., if you're plotting residuals.