We examined the distribution of 5,654 noncoding genome-wide significant associations (5,134 unique SNPs; fig. SI, table S2) for 207 diseases and 447 quantitative traits (2) with the deep genome-scale maps of regulatory DNA marked by DHSs. This revealed a collective 40% enrichment of GWAS SNPs in DHSs (fig. SIC, P< 10-55, binomial, compared to the distribution of HapMap SNPs). Fully 76.6% of all noncoding GWAS SNPs either lie within a DHS (57.1%, 2,931 SNPs) or are in complete linkage disequilibrium (LD) with SNPs in a nearby DHS (19.5%, 999 SNPs) (Fig. 1A) (12). To confirm this enrichment, we sampled variants from the 1000 Genomes Project (13) with the same genomic feature localization (intronic vs. intergenic), distance from the nearest transcriptional start site, and allele frequency in individuals of European ancestry. We confirmed significant enrichment both for SNPs within DHSs (P < 10-59, simulation) and also including variants in complete LD (r2=1) with SNPs in DHSs (P<10-37, simulation) (fig. S2).
