Genome-wideassociation studies frequently identify associations withmanyhighly correlated single-nucleotide polymorphisms (SNPs) in a chromosomal region, due in part to linkage disequilibrium, among the SNPs. This can make it difficult to determine which SNP within a group is likely to be the causative or functional variant. A, Genomic locations of 2 genes, the interleukin 23 receptor (IL23R) and the interleukin 12 receptor, beta-2 (IL12Rb2), and a hypothetical protein, NM_001013674, between positions 62700000 and 67580000 of the short arm of chromosome 1 at region 1p31, are shown. B, The −log10 P values for association with inflammatory bowel disease are plotted for each SNP genotyped in the region; those reaching a prespecified value of −log10 of 7 or greater are presumed to show association with disease. Several strong associations, at −log10 P values or greater, are seen in the region just telomeric of position approximately 67400000 and extending just centromeric of position approximate 67450000.

C, Pairwise linkage disequilibrium estimates between SNPs (measured as r2) are plotted for the region. Higher r2 values are indicated by darker shading. The region contains 4 “triangles” or “blocks” of linkage disequilibrium, 2 on either side of position 67400000 in the IL23R gene, another in the hypothetical protein telomeric of IL23R, and a fourth in the IL12RB2 gene at the centromeric end of the region. The 2 IL23R linkage disequilibrium regions each contain SNPs associated with inflammatory bowel disease, while the IL12RB2 region does not. Reproduced with permission from Duerr et al.53