Skip to main content
Fig. 7 | BMC Genomics

Fig. 7

From: Genome-wide enhancer annotations differ significantly in genomic distribution, evolution, and function

Fig. 7

Enhancers identified by multiple methods have little additional evidence of function. (a) Enrichment for overlap between conserved elements (n = 3,930,677) and liver enhancers stratified by the number of identification methods that predicted each enhancer. (b) Enrichment for overlap between GWAS SNPs (n = 20,458) and liver enhancers stratified by the number of identification methods that predicted each enhancer. (c) Enrichment for overlap between GTEx eQTL (n = 429,964) and liver enhancers stratified by the number of identification methods that predicted each enhancer. In (a-c), the average enrichment compared to 1000 random sets is plotted as a circle; error bars represent 95% confidence intervals; and n gives the number of enhancers in each bin. The only significant differences are found in the enrichment for evolutionary conservation (a), but the difference is modest in magnitude (1.36x for 1 vs. 1.62x for 6+). (d) Boxplots showing the distribution of confidence score ranks for FANTOM enhancers in liver partitioned into bins based on the number of other methods that also identify the region as an enhancer. Lower rank indicates higher confidence; note that the y-axis is flipped so the high confidence (low rank) regions are at the top. The lack of increase in enhancer score with the number of methods supporting it held across all methods tested (Additional file 1: Figure S14–17)

Back to article page