Figure 3
From: Genome-wide analysis of core promoter elements from conserved human and mouse orthologous pairs
The number of core promoter elements in the real promoter sequences (dark bars) significantly exceeds the numbers found in the randomized sequences (light bars) in all cases. When we add the criterion that the element must be conserved in the mouse genome, we find that the gap between the number of elements found in the real data versus the random data widens, indicating an increase in the signal-to-noise ratio.