First of all, we should understand that the input parameter accepted by limma is an expression matrix, and it is the expression matrix after log (with 2 as the bottom).

Then, the value of the last calculated logfc column is actually the value of the average expression of case group minus the average expression of control group in the input expression matrix. Then there will be positive and negative, which represents whether the gene is up-regulated or down regulated for case is equivalent to the control group.

I always had questions before. It’s obviously the difference between the average expression of case group and control group. It has nothing to do with log foldchange.

Later, I finally figured it out, because we input the expression matrix after log, so the average expression of case group and control group are log, and their difference is actually the foldchange of log

First of all, we need to understand the meaning of foldchange. If the average expression of case is 8 and the control is 2, then foldchange is 4 and logfc is 2

In the limma package, the average expression amount of case is 3 after log and control is 1. Then the difference is 2, that is, logfc is 2.

This is not a coincidence, but a very simple mathematical formula log (x / y) = log (x) – log (y)

Transferred from:http://www.bio-info-trainee.com/1209.html