Abstract
Background: Incidence of type 1 diabetes varies widely around the world, probably due to ethnic differences across populations among other factors.
Aims: To determine whether there is an association between disease and ancestry proportions; and to control disease–HLA associations for possible confounding by admixture or population stratification.
Subjects and methods: 100 cases and 129 controls participated in the study. Ancestry informative markers, which have considerable differences in frequency between European, West African and Native American populations were used. Type 1 diabetes associated HLA susceptibility/protection alleles were ascertained by PCR using specific primers. Statistical analyses were conducted using STRUCTURE 2.1, ADMIXMAP 3.7, SPSS 16.0 and STRAT 1.0 packages.
Results: The results of logistic regression implemented in ADMIXMAP 3.7 indicated that European ancestry was associated with type 1 diabetes mellitus with an odds ratio of 5.7 corresponding to one unit change in European admixture proportion. Association was found between HLA alleles and disease, DQA1*0501, *0301 DQB1*0201 and DRB1*0301, *0401 being susceptibility alleles and DRB1*1501, DQA1*0102/3 and DQB1*0602 being protective alleles.
Conclusions: We found an association between European ancestry and type 1 diabetes in our sample, indicating the contribution of ethnicity to incidence differences. Previously reported associations of HLA DR/DQ alleles with disease are confirmed for the admixed Cuban population.
Acknowledgements
We thank M. Hernández, S. Serrano, J. Clemente, J.C. Perez, M. Andreu and R. Echevarria for their technical assistance.
Declaration of interest: This work was supported in part by a grant from Ministry of Health, Vice-Ministry of Science and Technology. The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.