How do you use False Discovery Rates (FDR)?

The FDR (False Discovery Rate) is a statistical measure used to assess the significance (importance) given to the protein/pathways results. It is computed using the original p-values and it is necessary when a lot of tests are performed, usually hundreds of thousands of tests (depending on how many proteins/pathways are in the experiment).

The 5% FDR is a default threshold that is generally used to distinguish results that look more promising and therefore worth looking in more detail. 


Proteins/Pathways that have an FDR less than 0.05 are considered statistically significant but in the volcano plots (for both proteins and pathways), the y-axis represents the -log10(p-value). As such, we consider as significant those proteins/pathways that lie above the 5% FDR threshold.