| Popis: |
Chapter 1 serves as an introduction to the topic of this thesis, and briefly outlines the contents of each chapter. In Chapter 2 of this thesis, we used a large dataset (N = 43,889) from the Netherlands Twin Register (NTR) to assess prevalence and tracking coefficients across the largest part of the lifespan (ages 8 to 80), with varying distances between the baseline exercise measurement and follow-up (2 to 22 years). We assessed the prevalence and tracking of total weekly volume of exercise as well as of the volume of activities in six specific exercise domains: team-based, solitary, competitive, non-competitive, externally paced, and internally paced. In Chapter 3 of this thesis, we performed a systematic search and a meta-analysis of adult twin studies on regular voluntary exercise behavior, as well as three other physical activity phenotypes. Twin-studies are particularly well-suited to estimate the heritability of a trait by comparing the difference in within-pair resemblance of monozygotic (MZ) or dizygotic (DZ) twins. In Chapter 4 of this thesis, we additionally included parent-offspring and sibling correlations from family studies in our review of studies assessing the genetic and environmental contributions to regular voluntary exercise behavior, as twin studies rely on a number of assumptions. In Chapter 5 we addressed assortment in combination with a final major shortcoming of the twin- and family-studies included in chapters 3 and 4. These previous studies only estimated the effects of A, C and E, but could not simultaneously assess the effects of non-additive genetic factors (D). To achieve this, we used data from 50,690 adolescent and adult participants from 19,543 nuclear pedigrees in the NTR. All family-relations amongst these individuals and within these nuclear pedigrees were known, and this knowledge was leveraged by using the Mendel software package (Lange et al., 2013). In Chapter 6 of this thesis, we performed a GWAS of the total weekly volume of exercise, as well as a GWAS for activities in the team-based, solitary, competitive, non-competitive, and externally/internally paced domains as defined previously. We used a sample of the NTR (N = 14,626) for which genome-wide data in the form of millions of single nucleotide polymorphisms (SNPs) was available as well as survey data on exercise activities. In Chapter 7 we used a number of these designs to test the causal effects of the ‘big five’ personality traits (neuroticism, extraversion, agreeableness, openness and conscientiousness) on the total volume of exercise behavior, as well as on the volume of competitive/team and non-competitive/solitary exercise activities. First, we assessed the cross-sectional and longitudinal associations between the five personality traits and the various exercise behaviors. The main findings in this thesis are summarized in Chapter 8, followed by a general discussion on the genetic epidemiology of exercise, and my thoughts on future developments in this field. |
