Fig. 4

Motifs and interconnected clusters of promoters and enhancers. a, b Distributions of interaction for promoters (a) and enhancers (b). c–k Interaction motifs involving promoters (p) and/or enhancers (e) together with their occurrence among HiCap interactions and the numbers of unique promoters and enhancers within the motifs. l The most highly interconnected set of promoters, arranged by genomic coordinates and annotated with gene name. Promoters in green are all connected to each other whereas the ones in grey are missing some interactions. m Overlap between enhancer–enhancer interactions identified in HiCap and in-house Hi-C interactions. n Percentage of interactions supported with different read pair thresholds. o Mean degree for nodes (i.e., enhancers) in subnetworks of only enhancer nodes bound by the same TF or chromatin remodeling protein. Asterisks indicate significance (*P < 0.05, **P < 0.01, ***P < 0.001) computed against randomized enhancer node selections in the network. p Number of significant gene ontology (GO) categories enriched among gene pairs that share the same category, for genes connected by HiCap interaction motifs. FDR false discovery rate. q Mean read pair support for promoter pairs with increasing number of common enhancer interactions (0, 1 or ≥2)