The proposed study will examine the relationship between genetics and environment on exercise capacity, a known indicator of cardiovascular and overall health.
Aim 1. Generate a comprehensive null model of exercise capacity. Specifically, develop a regression model that best explains the outcome of exercise capacity using design variables and covariates without strong genetic influence, such as BMI. The model will then be extended to incorporate genetic information (i.e. SNP data).
Aim 2. Generate activity signatures using accelerometer data.
Aim 3. GWAS of exercise capacity. Knowledge about how the combination of genetic variants and physical activity patterns affect exercise capacity will support further translational research. Identifying patterns of physical activity that increase exercise capacity will potentially determine valuable guidelines for providing individuals with feedback on lifestyle changes that may prove beneficial for their overall health. We will use regression techniques to select the most informative conventional predictors of exercise capacity (such as heart rate). In addition, we will apply machine-learning techniques to accelerometry data in order to generate hidden activity signatures. The combination of conventional variables such as smoking and analytically defined activity signatures will then be correlated with genetic variants through GWAS. Full cohort