CEH frequency in African, European, and African American control populations.
African control frequencies are from Gragert, et al. [26], European control frequencies are from the WTCCC [13, 14], and AA control frequencies are from the AA dataset reported here. African CEHs are highlighted in orange whereas European CEHs are highlighted in yellow (see Text). Only CEHs with ≥12 representations in the AA cohort are listed. The “No SNP haplotypes” condition is for CEHs not including any associated SNP-haplotype (Tables 1 & 2). The “With SNP haplotypes” condition is for CEHs that include the associated SNP-haplotype as indicated in Tables 1 & 2. Name indicates the haplotype (Tables 1 & 2), sorted in descending order of frequency in the WTCCC [13, 29]–designated by (c)–and in the AA cohort for CEHs not found in the WTCCC–designated by (b). In the Table, (0) indicates true zero. Ratios are of CEH frequency in Europeans to that in Africans (No SNPs condition) and the frequency ratio in Europeans to that in African Americans (With SNPs condition). nd = not defined; ∞ = infinity. Significance of the difference in CEH frequency between Africans and Europeans are indicated as follows:
* * 10−4 ≤ p < 0.05
*** 10−8 ≤ p < 10−4
†* * 10−12 ≤ p < 10−8
††* p < 10−12
It appears that the Class II haplotypes DRB1*07:01_DQB1*02:02 and DRB1*09:01_DQB1*02:02 in the WTCCC data are consistently identified in both African and European populations as DRB1*07:01_DQB1*02:01 and DRB1*09:01_DQB1*02:01, respectively, in the data of Gragert and colleagues [28]–see Text. If these Class II haplotypes are, in fact, the same, there is still a significantly greater (c7) CEH frequency in Europeans compared to Africans (p<10−8). However, in this case, the (b14) CEH occurs only in Africans, whereas the (b10) CEH still occurs in neither Africans nor Europeans.* * 10−4 ≤ p < 0.05
*** 10−8 ≤ p < 10−4
†* * 10−12 ≤ p < 10−8
††* p < 10−12
