Third, at a given GWAS locus, the SNP with the most significant association (that is, the lowest P value) with the disease is usually reported as the ‘lead’ SNP. Except in rare instances, such as the SNP rs6983267, which influences the MYC enhancer and confers risk for multiple cancers [34,35], the SNP with the lowest P value is not necessarily causal. Any SNP in LD with the lead SNP may be causal, and there may be dozens to thousands of candidates. Fine mapping studies can help narrow the locus and reduce the number of candidates. Additionally, as discussed above, identifying SNPs that co-localize with enhancer-chromatin features or TFBSs in an appropriate human cell type can help prioritize candidate functional variants [30,38]. Indeed, Claussnitzer and colleagues developed a method, phylogenetic module complexity analysis (PMCA), which utilizes conserved co-occurring TFBS patterns to identify functional regulatory variants [82]. However, hundreds of candidate SNPs may remain even after prioritization, especially when the locus harbors an enhancer cluster. This was illustrated in a recent survey of breast cancer risk loci, which showed that 921 SNPs
