TABLE 2

Common haplotypes at IRS1 and their association with type 2 diabetes in Pima Indians

HaplotypeCompositionFrequencyFrequency (diabetes)Frequency (nondiabetic)Odds ratio (95% CI)P
IG_C_G_A0.390.380.420.78 (0.62–0.97)0.15
IIG_C_G_G0.380.380.380.97 (0.78–1.20)1.00
IIIA_T_A_G0.100.100.101.00 (0.71–1.41)1.00
IVA_C_A_G0.050.060.032.44 (1.46–4.06)<0.01
VA_C_G_A0.030.030.041.16 (0.65–2.07)0.99
VIA_C_G_G0.020.020.021.75 (0.79–3.92)0.67
VIIG_T_A_G0.010.020.012.44 (1.04–5.71)0.22
  • A modification of the zero-recombinant haplotyping method was used to assess association between traits of interest and haplotypes of the four IRS1 SNPs (23,24). Haplotypes are defined by the composition of alleles at each SNP in following order: −7120G/A_−2223C/T_−446G/A_Ala804. Odds ratios are calculated adjusted for age and sex for a difference of one in the number of the relevant haplotype using a binomial generalized estimating equation model that accounts for family membership; P values are adjusted for the multiple comparisons involved in analysis of 7 different haplotypes using the Bonferroni correction [corrected P value = 1 − (1 − P)6].