TABLE 2

Frequencies of the major PTPN1 haplotypes in the various diabetic subsamples

StudyControl
Type 2 diabetes
SampleAB + DSampleAB + D
Bento et al. 1 (13)Simultaneously ascertained0.450.33Type 2 diabetes and end-stage renal disease0.410.39
Bento et al. 2 (13)Public repository0.470.34Type 2 diabetes; obese (mean BMI 33.7 kg/m2)0.390.41
Palmer et al. (14)Urban and rural Hispanics0.430.39
Average frequency0.450.35Average frequency0.400.40
This studyGCI U.S. control0.440.35Individually matched type 2 diabetes0.440.33
This studyGCI Poland control0.400.38Individually matched type 2 diabetes0.420.37
This studyCanadian control0.460.34Individually matched type 2 diabetes0.500.29
This studyScandinavian control0.450.37Individually matched type 2 diabetes0.480.34
This studyScandinavian sibling0.460.36Scandinavian sibling with type 2 diabetes0.470.33
This studySwedish control0.460.34Individually matched type 2 diabetes0.430.36
Average frequency0.450.36Average frequency0.460.34
  • The frequencies of the two major haplotypes described in Bento et al. (13) (haplotype A and the combination of haplotypes B and D in this study) were measured in the control and diabetic samples of Bento et al. (13), Palmer et al. (14), and the various subsamples examined in this study. The GCI case/control samples were individually matched for age, sex, and self-reported ethnicity, whereas the Canadian and Scandinavian case/control samples were individually matched for age, sex, and BMI. Haplotypes in this study were derived by phasing all case and control chromosomes together using an expectation maximization algorithm (see text for details).