Fig. 1
From: The GC-content at the 5′ ends of human protein-coding genes is undergoing mutational decay
Nucleotide content of genomic regions surrounding TSS sites of human protein-coding genes. For all annotated human protein-coding genes (N = 18,874), the average GC-content (A–C), nucleotide content on the coding strand (D–F), and CpG-content (G–I) were plotted (y-axis) against the nucleotide position surrounding the TSS (x-axis; 2 kb surrounding the TSS in A, D, and G; 10 kb surrounding the TSS in B, E, and H) with negative numbers indicating upstream sequence and positive numbers indicating downstream sequence, or along binned genomic sequence the TSSs and first exon-intron boundaries (EIB) of all human protein-coding genes were aligned so that each gene sequence was “normalized” for the length of the first exon (x-axis C, F, and I)