Figure 1. Mean GC content of retained introns, exons bordering retained introns and non-retained introns in human AS-NMD candidates.

The data sets was divided into five categories to avoid bias due to their length. The data set of ‘Non-retained introns’ was generated by randomly picking a non-retained intron from the set of genes that contain retained introns, so that the same number of sequences is in each bin. In addition, we included bars in the plot for 'Retained introns that are conserved between human and mouse.