Figure 6

SNPs discovered from ChIP-seq and disease-associated SNPs. (A) The proportions (per 10,000) of SNPs that were identified either by the 1000 Genomes Project Pilot 2 set (1000 Genomes trio) or discovered from ChIP-seq data in this study (ChIP-seq), that were also associated with diseases based on the NHGRI GWAS catalog (http://www.genome.gov/gwastudies) are shown. ChIP-seq Set 1 refers to those SNPs discovered de novo from ChIP-seq data that were also called by the 1000 Genomes Pilot 2, to allow direct comparison. ChIP-seq Set 2 refers to SNPs called from ChIP-seq data generated by us in cell lines that were not genotyped by the 1000 Genomes Project. ChIP-seq Set 3 refers to SNPs called from ChIP-seq data from other ENCODE labs in cell lines that were not genotyped by the 1000 Genomes Project. (B) Allele-specific binding near GWAS SNPs. Each point represents a significantly biased allele-specific binding event of CTCF at the indicated distance from a GWAS locus. Shown here are all allele-specific binding loci within +/− 500 bp of a GWAS site (which would be at position 0). Multiple points at the same distance represent allele-specific binding in different individuals. (C) An example from the set of allele-specific binding sites near a GWAS SNP. The two sites indicated by red dots in 'B' correspond to the heterozygous sites in 'C'. The read density of ChIP-seq data at the binding site is shown in selected individuals. When heterozygous, the pie chart indicates relative occupancy of the two alleles.