Extended Data Fig. 14: Chromatin state-based enhancers. | Nature

Extended Data Fig. 14: Chromatin state-based enhancers.

From: An atlas of dynamic chromatin landscapes in mouse fetal development

Extended Data Fig. 14

a, Tissue-specific enrichments of VISTA enhancers for different chromatin states in E11.5 heart, limb and forebrain. b, Top, fraction of dynamic enhancers in each tissue (based on H3K27ac) that overlap d-TACs accessible in the matching tissue. Bottom, fraction of dynamic enhancers in each tissue that overlap d-TACs that were also called as dynamic by ATAC–seq in the matching tissue. c, Top, fraction of dynamic d-TACs in each tissue that overlap enhancers called by ChromHMM (state 5) in the matching tissue. Bottom, fraction of dynamic d-TACs in each tissue that overlap dynamic enhancers called with H3K27ac in the matching tissue. Each point represents one tissue-stage (n = 66). d, Top, dynamic enhancers that gain H3K27ac at a given stage transition n to n + 1. Lines show the log2 fold change in ATAC–seq signal within d-TACs that overlap those dynamic enhancers at various stage transitions. Dynamic enhancers that gain H3K27ac at a given stage transition tend to gain accessibility as measured by ATAC–seq either at or before the stage transition in question (sometimes preceding H3K27ac gain by as much as five stage transitions). Mean and s.d., filled circles and vertical lines, respectively. Bottom, dynamic enhancers that lose H3K27ac at a given stage transition n to n + 1. Dynamic enhancers that lose H3K27ac at a given stage transition tend to lose accessibility as measured by ATAC–seq either at or after the stage transition in question (sometimes proceeding H3K27ac loss by as much as five stage transitions). The number of stage comparisons for each offset is: ±0 n = 54, ±1 n = 42, ±2 n = 30, ±3 n = 18, ±4 n = 10, ±5 n = 5.

Back to article page