Figure 3

Schematic of the scanning process to find regions with promoter patterns. Scan genome using the fitted models to find correlated regions. Gray area (top) is the ChIP-seq data of Pol-II and gray area (bottom) is the ChIP-seq data of H3K4me2. Red curve (top, red-solid-curve) is the double exponential-uniform model fitted to the Pol-II patterns and blue curve (bottom, blue-dot-curve) is the double exponential-uniform model fitted to the H3K4me2 patterns. Correlated regions are identified as region which have significantly high correlation coefficient with both Pol-II and H3K4me2 models (curves).