Supplementary Figure 12 : CAPRA use cases

Supplementary Figure 12 | CAPRA use cases

a, Density scatterplot showing the correlation of CAPRA-calculated STARR scores for promoter cCREs in K562 (x-axis) and HepG2 (y-axis). b, Density scatterplot showing the correlation of STARR scores for distal enhancer cCREs in K562 (x-axis) and HepG2 (y-axis). These data are also presented in Figure 2a but are colored here by point density rather than cell type specificity. c, Principal component analysis (PCA) of STARR scores across 362,591 cCREs from seven STARR-seq libraries. Each point represents a library, with color indicating cell type and shape indicating lab origin (triangles for Reddy lab, circles for White lab). PC1 and PC2 explain 32% and 28% of the variance, respectively. d, Heatmap of STARR scores for the top 500 cCREs contributing to PC1 from panel c. Three major clusters are shown, identified by hierarchical clustering; enriched sequence motifs for each cluster are listed at right. e, Heatmap of STARR scores for the top 500 cCREs contributing to PC2 from panel c. Two major clusters are shown, with corresponding enriched motifs listed at right. f, Line plots showing RNA:DNA ratios for the cCRE pair EH38E2809119 and EH38E2809120 using CAPRA’s Paired Sweep mode. Gray and black lines represent fragments anchored at EH38E2809119 and EH38E2809120, respectively; line width is proportional to the number of input DNA fragments. FIMO-identified motifs across the region are shown below the plots, with four highlighted motif groups: canonical E-box (orange), high GC content (red), non-canonical E-box (green), and ZNF/STAT motifs (blue).