Article
Genomewide Association Study of Body Mass Index, waist circumference and waist hip ratio in Scandinavian individuals with diabetes or with normale glucose tolernace
Search Medline for
Authors
Published: | September 6, 2007 |
---|
Outline
Text
Obesity and its complications, including type 2 diabetes, have reached epidemic proportions. Body mass index (BMI), waist circumference (WC) and waist hip ratio (WHR) are heritable measures of obesity and predict future risk of diabetes, cardiovascular disease, and death. We hypothesize that common genetic variants influence BMI, WC, and WHR. We applied the Affymetrix 500K platform to DNA samples individuals from Sweden and Finland that were part of a case/control study of diabetes (Diabetes Genetics Initiative) matched for age, gender and BMI. The median and interquartile range of BMI, WC, and WHR in cases was significantly higher than in controls (Table 1). For association analysis, we log transformed BMI, WC, and WHR and then calculated age and gender-adjusted z-scores, using population based data. From the Affy 500K platform, 389,869 SNPs passed quality control filters (genotyping success in >95% individuals using the BRLMM algorithm, minor allele frequency >0.01, HWE p>10 -6). We tested SNPs for association under an additive genetic model using the PLINK software package. We observed an excess of low p values (Table 1) but overall inflation factors were close to one, indicating that the study was not substantially affected by systematic technical biases such as population stratification. These results suggest that common genetic variants may influence BMI, WC, and WHR in this population, but modest effect sizes will necessitate additional studies in order to distinguish which preliminary results represent false positives due to statistical fluctuations and which represent true associations with BMI, WC and WHR. We are testing the SNPs with the best evidence of association in independent