Gene-lifestyle interactions of relevance to type 2 diabetes have been difficult to identify but may help stratify groups of people at especially high risk. In large genetic studies, measures of diabetes risk factors such as diet and activity are usually limited to self-reported data. We used genetic and accelerometer data from 97,8individuals to test for genetic interactions with physical activity and sleep lifestyles. We tested the hypothesis that lower levels of physical activity and abnormal sleep patterns accentuate genetic susceptibility to obesity. We used 82,322 individuals from the UK Biobank, 10,885 from the Women’s Genome Health Study (WGHS) and 4,599 individuals from the Avon Longitudinal Study of Parents and Children (ALSPAC). We derived measures for physical activity and sleep, including total physical activity, time spent undertaking sedentary, light, moderate and vigorous activity, and sleep duration and sleep efficiency. We used BMI as the outcome and tested for interactions between a BMI genetic risk score (GRS) and the accelerometer derived measures of activity and sleep. We also performed several negative control experiments to test for residual confounding. We found evidence of gene-activity interactions. For example, in the UK Biobank, when analysing non-sedentary physical activity, the 10% of individuals carrying the most BMI-raising alleles were 5.9kg heavier if they were inactive compared to active (for height 1.73m). In contrast, the 10% of individuals carrying the fewest BMI-raising alleles were 4.1kg heavier if they were inactive compared to active (Pinteraction=4×10-4). This observation was consistent in the WGHS and ALSPAC data although not all tests reached P<0.05. The results were robust to negative control tests, including the testing of interactions using random variables simulating the BMI-activity correlations. Our results are consistent with previous studies suggesting that low levels of physical activity accentuate the genetic risk of obesity.


A.R. Wood: None. S.E. Jones: None. R. Richmond: None. S. Ahmad: None. H. Yaghootkar: None. R. Beaumont: None. K.S. Ruth: None. M. Tuke: None. A. Murray: None. R.M. Freathy: None. G. Davey Smith: None. I. Lee: None. N. Timpson: None. D. Chasman: None. Z. Kutalik: None. M.N. Weedon: None. J. Tyrrell: None. T.M. Frayling: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at