Utility of large consanguineous family-based model for investigating the genetics of type 2 diabetes mellitus

Objectives: This study examined the utility of a family-based model for replicating the results of genome-wide association studies (GWAS) of type 2 diabetes (T2D). Methods and results: In a total of 232 members of a large consanguineous Omani Arab pedigree (age: 16-80. years), there were 27 diabetics and 50 prediabetics (17 with impaired fasting glucose and 33 with impaired glucose tolerance). All 232 individuals underwent anthropometric and biochemical investigations and genotyped for 14 known common gene variants of modest effect on T2D risk. Power analysis at a LOD score of 3, gave 80% power to locate a single specific locus that accounts for 52% of the total phenotypic variation. Measured genotype analysis (MGA) was used to determine heritability of various quantitative traits (QTs) which ranged 25-56%. Using MGA, some common gene variants were found to have little (< 5%) but significant impact on the heritability of T2D related QTs [. KCNJ11 (rs5219), p = 0.004]; [. IGF2BP2 (rs4402960), p = 0.02]; [. SLC30A8 (rs13266634), p = 0.05]; [. CAPN10 (rs2975760), p = 0.031]; [. FTO (rs8050136), p = 0.023]; [. FTO (rs9939609), p = 0.018] and [. SLC30A8 (rs13266634), p = 0.05]. Sib-TDT analysis showed that some gene variants were significantly associated with T2D risk but didn't reach the level of significance after Bonferroni correction [. KCNJ11 (rs5219), p = 0.047] and [. CAPN10 (rs41266971), p = 0.035]. Conclusion: We have demonstrated that, in principle, a family-based model with minor limitations could be used to replicate some of the results of large GWAS case-control studies. This model could successfully be applied for the future discovery, by deep sequencing, of rare gene variants.