Fig. 6
From: Origin of exon skipping-rich transcriptomes in animals driven by evolution of gene architecture
Representative patterns of the inter-relationships between GC content and AS. For selected eukaryotes (a–e), we plotted the distribution of GC content (percentage) in long and short introns (first row); the distribution of GC content in introns sorted according to their length (five quantiles, Q1–5), and ES status (negative and positive; second row); and the differential between intron and exon GC content in long introns (Q5) of ES-positive and ES-negative events, using box plots (third row) and empirical cumulative distribution functions (fourth row). In the second row, we used Spearman’s rank correlation coefficient to test the association between intron length and GC content (significant at p < 0.01). In the third and fourth rows, we used Kolmogorov-Smirnov one-sided test to test whether ES-positive introns had higher ΔGCin–ex than ES-negative ones (i.e., ΔGCin–ex ~ 0), and thus GC-homogeneous across introns and exons (significant p < 0.01)