Heritability of physical activity traits in Brazilian families: the Baependi Heart Study
© Horimoto et al; licensee BioMed Central Ltd. 2011
Received: 27 May 2011
Accepted: 29 November 2011
Published: 29 November 2011
It is commonly recognized that physical activity has familial aggregation; however, the genetic influences on physical activity phenotypes are not well characterized. This study aimed to (1) estimate the heritability of physical activity traits in Brazilian families; and (2) investigate whether genetic and environmental variance components contribute differently to the expression of these phenotypes in males and females.
The sample that constitutes the Baependi Heart Study is comprised of 1,693 individuals in 95 Brazilian families. The phenotypes were self-reported in a questionnaire based on the WHO-MONICA instrument. Variance component approaches, implemented in the SOLAR (Sequential Oligogenic Linkage Analysis Routines) computer package, were applied to estimate the heritability and to evaluate the heterogeneity of variance components by gender on the studied phenotypes.
The heritability estimates were intermediate (35%) for weekly physical activity among non-sedentary subjects (weekly PA_NS), and low (9-14%) for sedentarism, weekly physical activity (weekly PA), and level of daily physical activity (daily PA). Significant evidence for heterogeneity in variance components by gender was observed for the sedentarism and weekly PA phenotypes. No significant gender differences in genetic or environmental variance components were observed for the weekly PA_NS trait. The daily PA phenotype was predominantly influenced by environmental factors, with larger effects in males than in females.
Heritability estimates for physical activity phenotypes in this sample of the Brazilian population were significant in both males and females, and varied from low to intermediate magnitude. Significant evidence for heterogeneity in variance components by gender was observed. These data add to the knowledge of the physical activity traits in the Brazilian study population, and are concordant with the notion of significant biological determination in active behavior.
Physical inactivity has been identified as the fourth leading risk factor for global mortality (6% of deaths worldwide), preceded only by high blood pressure (13%), tobacco use (9%) and high blood glucose (6%). Inactivity is associated with several diseases, such as cardiovascular disease, type 2 diabetes, obesity, hypertension, colon and breast cancer, anxiety and depression .
Numerous epidemiologic studies have emphasized the importance of regular physical activity in the prevention and treatment of several common chronic diseases. Despite the benefits of physical activity on health, levels of inactivity are high in virtually all developed and developing countries. At least 60% of the world population presents a sedentary behavior, mainly because of insufficient participation in physical activity during leisure time and a decrease in occupational and domestic activities . In Brazil, from 10.7% to 22.1% of the adult population was classified as completely inactive, with indexes higher in males (15.0%) than in females (13.6%); the maximum frequency of inactivity was in the age group over 65 years old (35.8% in males and 37.7% in females) .
Participation in physical activity is a behavioral trait determined by environmental and genetic factors that may reflect, in part, the predisposition to adopt and maintain a physically active lifestyle. Among adults, physical activity includes leisure time physical activity, transportation (e.g. walking or cycling), occupational activity (i.e. work), household chores, and participation in games, sports or planned exercise, all in the context of daily, family, and community activities. There are specific recommendations of physical activity for healthy adults in the age ranges of 18-65 years and 65 years and above to promote and maintain health which take into account the quantity (in minutes) and the intensity of the activity [2, 4].
Familial aggregation of distinct physical activity dimensions, such as total or weekly physical activity, leisure time activity, and level of daily physical activity [5, 6] has been suggested mainly from twin and family studies. Diversity in sample size, age, gender, phenotype definition, and statistical procedures among the different studies probably explains why heritability estimates have ranged from 30% to 83% in twin studies and from 10% to 30% in studies with nuclear families . Nonetheless, heritability accounts for a substantial portion of the variation of physical activity phenotypes that have been reported at the population level.
In this study, we estimated the genetic influences on sedentarism, weekly physical activity (exclusively related to participation in sports activities), weekly physical activity among non-sedentary subjects, and level of daily physical activity (related to work and free-time activities, excluding sports) in the families that participated in the Baependi Heart Study. We also investigated whether genetic and environmental components contribute differently to the expression of these phenotypes in males and females.
The Baependi Heart Study
The Baependi Heart Study  is a genetic epidemiological study of cardiovascular disease risk factors, with a longitudinal design. Baseline enrollment occurred between December 2005 and January 2006 when 1,857 individuals distributed in 95 families resident in the municipality of Baependi, a city located in the Southeast of Brazil were selected to participate in the study. Probands were identified, in several stages, from the community at large. Eleven census districts (from a total of twelve) were selected for the study and the residential addresses within each district were randomly selected (first by randomly selecting a street and then a household). Only subjects age 18 and older, living in the selected household, were eligible to participate in the study.
When a proband was enrolled, all his/her first-degree (parents, siblings, and offspring), second-degree (half-siblings, grandparents/grandchildren, uncles/aunts, nephews/nieces, and double cousins), and third-degree (first-cousins, great-uncles/great-aunts, and great-nephews/great-nieces) relatives and the relatives of their respective spouse's who were at least 18 years old, were invited to participate. After the first contact with the proband, the first degree relatives were invited by phone to participate; all living relatives in the city of Baependi (urban and rural area) and surrounding cities were included in the study. To recruit the participants, the study was advertised through the provincial, religious, and municipal authorities, in local television, newspaper, and radio messages, through physicians, and by phone calls. For the physical examination, a clinic was established in an easily accessible sector of Baependi.
Information regarding family relationships, sociodemographic characteristics, medical history, and environmental risk factors such as physical activity, smoking habits, and alcohol use were evaluated through a questionnaire that was applied to each participant. The questionnaire was based on the WHO-MONICA epidemiological instrument, which was applied and filled out by research assistants specially trained for this task. Anthropometric measures such as weight and height were measured following the standardized procedures described earlier  and they were employed to define the body mass index (BMI) (kg/meters2).
The study protocol was approved by the ethics committee of the Hospital das Clínicas, University of São Paulo, Brazil, and each subject provided informed written consent before participation.
Physical activity profiling
The physical activity profile of this population was delineated using four dimensions: sedentarism; weekly physical activity exclusively related to participation in sports activities (weekly PA); weekly physical activity among non-sedentary subjects only (weekly PA_NS); and the level of daily physical activity, including work activities (daily PA). Three global questions were used for data collection.
The first question was applied to assess the daily level of physical activity related to work and free-time activities, excluding sports. Four choices were presented, considering actions such as sitting, standing, walking, and object lifting in daily activities: (1) Do you have to remain seated to perform your activities? Do you walk while working?; (2) Do you walk a lot while performing your activities but do not need to lift or carry heavy objects?; (3) Do you walk and move many objects or go up and down stairs or an incline?; (4) Do your activities require major physical effort, for instance, do you lift or cut heavy objects? These choices represent levels of physical effort; extremely light (1), light (2), moderate (3), and vigorous (4). The subjects self-rated their daily level of physical activity, choosing the option that best represented their daily actions.
The second question was related to the practice of sports activity and its frequency: Do you take part in sports? In the case of an affirmative answer, how many days a week do you participate? Eight choices were presented: (1) I do not take part in sports; or I take part in sports (2) once a week; (3) twice a week; (4) three times a week; (5) four times a week; (6) five times a week; (7) six times a week; or (8) seven days a week. Subjects that chose the first option were classified as sedentary and the others, as active. This information was used to compose the sedentarism phenotype.
The last question registered the average time (in minutes) of each sport session for those subjects that participated in sports: What is the average duration of each sports session? A weekly PA phenotype was computed by multiplying the number of days a week in which the individual engaged in sports, assessed by the previous question, and the average time of each session.
Among the 1,857 individuals selected in the baseline phase of the Baependi Heart Study, 1,693 (98.9%) answered the questions that define the sedentarism and weekly PA phenotypes, and 1,675 (97.8%) answered the related question about daily physical activity. The weekly PA_NS data set is a subgroup of the weekly PA sample, which was composed of 515 non-sedentary subjects. Of the subjects that were classified as having the sedentarism phenotype because they did not practice sports, 283 subjects were engaged in work or free-time activities that required major physical effort (moderate or vigorous daily activities); they were therefore excluded from the data set to avoid misclassification. So, the final sample for the sedentarism analysis was composed of 1,410 subjects, 873 defined as sedentary and 537 defined as active.
The sedentarism phenotype was analyzed as a dichotomous variable contrasting sedentary versus active subjects. The other traits were analyzed as continuous variables. Natural log-transformation was applied for weekly PA and weekly PA_NS traits to achieve the required normality assumption, followed by a new data assessment.
Familial correlations using the pairwise weighting scheme were computed using the FCOR program within the SAGE software package  for all main pair types of relatives available in the pedigrees.
Polygenic heritability estimates were calculated for all traits using the variance-components approach implemented in the SOLAR package . In the most narrow sense, the heritability of a trait represents the proportion of the phenotypic variance attributable to additive genetic effects and is given by h2 = σ2 a/σ2 p, where σ2 a is the variance due to the additive effects of genes and σ2 p is the phenotypic variance. For quantitative traits, the overall phenotypic variance was estimated from the observed distribution of trait values in the sample, and was partitioned into genetic and environmental components using the observed covariance among family members, as Ω = 2Φσ2 a + Iσ2 e, where Ω is an nxn matrix of the n individuals in the data set; 2Φ is the structuring matrix of the coefficient of relationship; and I is an identity matrix that represents the structuring matrix for σ2 e, the variance due to residual environmental factors. Covariates were not dealt with as a variance component, but rather as a modification to the trait mean, and covariate-specific trait means were used in the calculation of covariance among relatives. An extension of this basic model, the liability threshold model, was used to analyze discrete or categorical traits .
Household group analyses were also performed using the SOLAR system . An additional variance parameter was added to model the effect of common environment, which is associated with any non-genetic factors shared between the individuals living in the same household at the time of study. So, the covariance matrix for pedigree, described above, was rewritten as Ω = 2Φσ2 a + Hσ2 c +Iσ2 e, where H is the structuring matrix for σ2 c, the variance due to common environment effects; this matrix could contain zeros and ones depending upon whether an environment factor was shared or not between each pair of individuals. Using current residential addresses to define households, we obtained 740 nuclear families from the 95 families that are part of the Baependi Heart Study. Household effects were investigated in all polygenic models studied.
The SOLAR system handles categorical variables with more than two classes as quantitative traits. Thus, we applied a robust estimation approach implemented in SOLAR through the tdist procedure, to correct for bias of the heritability estimate of the daily PA phenotype.
Two models were fitted to the data: considering no covariate effects (model I); and considering age, sex, age2, and age by sex interaction effects, simultaneously (model II). Covariate screening determined the statistical significance of each effect; covariates with low significance levels (P ≥ 0.1) were removed from the final model, and the variance caused by all remaining covariates was computed. Household effects were included in all models.
Models with distinct genetic and environmental variance components were also employed to assess the evidence of heterogeneity among genders in the heritability estimates of the studied traits following the method described by Giolo et al. . Assuming that the phenotypes in males and females are influenced by the same set of genes with distinct effects among genders, four situations regarding the genetic and environmental variance components were considered: homogeneity in both variance components, heterogeneity in at least one of the variance components, heterogeneity only in the environmental variance components, and heterogeneity only in the genetic variance components. Likelihood ratio tests were applied to define the models that presented the best fit to the data.
Models with distinct genetic and environmental variance components between the genders did not adequately fit to the daily PA dimension, even after a robust estimation approach implemented in SOLAR (tdist procedure) was employed to try to correct the bias of the heritability estimates. For this phenotype, the heritability estimates for the genders were obtained by considering males and females in separate analyses. Polygenic heritability estimates were calculated from model I (no covariate effects) and model II (with covariate effects) with the heritability estimation considering no heterogeneity among genders.
Anthropometric and sociodemographic characteristics of the study population
Men (n = 734)
Women (n = 959)
Total (n = 1,693)
BMI (kg/m 2 )
Education level (%)
≤ 8 years (low)
9-11 years (moderate)
≥ 12 years (high)
Socioeconomic status (%)
Class A + B (high)
Class C (moderate)
Class D + E (low)
Distribution of phenotypes according to gender and age groups in the study population
Interclass and intraclass correlations in different pairs of relatives for the different phenotypes
Two polygenic models, model I (no covariate effects) and model II (with covariate effects) were fitted to the data in all the analyses. Only model I fitted the weekly PA_NS data. Model II was also applied to the weekly PA_NS data, but the covariates presented no significant effect on this phenotype. We did not observe significant household effects for any of the studied traits, irrespective of the model used for the analysis.
Parameter estimates for the sedentarism, weekly PA, and weekly PA_NS phenotypes in the Brazilian families
Heterogeneity in variance components by gender
Variance with heterogeneity
h2 (s.e.), p value
σ2 g, m
σ2 e, m
σ2 g, f
σ2 e, f
0.14 (0.07), p = 0.0104
Genetic and environmental
age × sex, age2, age2 × sex
0.12 (0.07), p = 0.0218
0.09 (0.04), p = 0.0019
Age, age × sex, age2, age2 × sex
0.11 (0.04), p = 0.0004
0.35 (0.11), p < 0.0001
There was no heterogeneity in the variance components by gender for the sedentarism, weekly PA, weekly PA_NS phenotypes when the models were adjusted by covariates. Using the model with no covariate effects, heterogeneity by gender was observed for the sedentarism trait in both (genetic and environmental) variance components, with heritability higher in females (0.22) than in males (0.05). For the weekly PA trait, heterogeneity by gender was observed in only the environmental variance component (heritability estimates were 0.09 in females and 0.11 in males). Because the models with distinct genetic and environmental variance components among the genders did not fit adequately to the daily PA phenotype, the heritability estimates were obtained from data sets that were separated for males and females, considering models I (no covariate effects) and II (with covariate effects). In males (727 subjects), the heritability estimates were 0.22 ± 0.07 and 0.21 ± 0.07 for models I and II respectively. The age2 covariate represented about 2% of the phenotypic variance of this trait, and its inclusion into the model slightly decreased the heritability estimate, following the trend observed in the analysis with no heterogeneity among genders. In females (947 subjects), there were no significant covariate effects for the daily PA trait and the heritability estimate was 0.11 ± 0.06. For model I, the heritability estimate obtained for males (22%) was exactly twice that for females (11%). There were no significant household effects for the daily PA phenotype in these analyses.
Heritability estimates for the daily PA phenotype from analysis with separate data sets by gender
Heterogeneity by gender
h2 (s.e.), p value
h2 (s.e.), p value
0.12 (0.04), p = 0.0001
0.22 (0.07), p = 0.0002
Sex, age × sex, age2
0.10 (0.04), p = 0.0005
0.21 (0.07), p = 0.0003
0.12 (0.04), p = 0.0001
0.11 (0.06), p = 0.0204
Sex, age × sex, age2
0.10 (0.04), p = 0.0005
In this study, the physical activity profile of the sample of Brazilian adults in the Baependi Heart Study was investigated using data collected by questionnaires based on the WHO-MONICA epidemiological instrument. Baependi is located in a rural area of Brazil and its economy is based on agriculture and this, for the most part, explains the low sociodemographic indicators observed in the sample and described in Table 1. Although the obesity prevalence was similar to the 2006 rate observed in the Brazilian population as a whole , the overweight prevalence in the sample was lower (26% in the study population versus 34.1-48.3% in the Brazilian population). Obesity is a complex disease determined by genetic predisposition and environment factors such as diet and sedentary behavior. Considering the low level of weekly physical activity observed in this sample, the lower overweight rate is probably associated with the level of daily activities and a healthy diet that is commonly observed among the inhabitants of rural areas. Subjects predominantly classified their daily activities as light, but the concept of light work in rural and urban areas is very different. Field work demands high physical effort and this probably explains the low overweight prevalence that was observed. In general, the level of weekly physical activity was higher in males than females and this declines with increasing age, consistent with previous studies  (Table 2).
Significant intraclass and interclass correlations were observed for sibling and avuncular relationships. This result indicated that environmental factors and lifestyles shared by persons of the same generation who are genetically and closely related play a significant role in the phenotypic variance of physical activity. Significant correlations among siblings for physical activity phenotypes have also been found in other populations, although the results are not directly comparable because of divergence in the studied phenotypes and in sample characteristics [15, 16]. One of these studies did assess the avuncular interclass correlation, but it was not found to be significant in any of the studied phenotypes . We have yet to explain the shared environment effects observed for the avuncular correlation in the weekly PA_NS phenotype. In our study, no significant parent-offspring correlation was found, although similarity in their physical activity habits was commonly detected. This result indicated that either different genetic factors were modulating physical behavior in the two generations, or that there were generation-specific environmental factors promoting more resemblance within generations than across generations . Another possible explanation is that of a drastic change in the rural Brazilian society occurred during the last decades and this may have diluted the effects.
Several approaches have been used to model the genetic contributions to physical activity phenotypes. Twin [18, 19] and family [20, 21] studies have been employed to dimension genetic factors, associations [7, 22, 23], linkages [23–25], and animal [26, 27] studies have contributed to the identification of candidate genes and metabolic alterations involved in the determination of physical activity phenotypes. Animal studies have also been used to describe the associations between physical activity phenotypes and other risk factors for cardiovascular diseases, such as obesity and type 2 diabetes. Different genomic regions have shown linkages with distinct activity phenotypes, suggesting that each of the phenotypes might be influenced by specific mechanistic pathways. In human, significant linkage signals have been identified on chromosomes 2p22-p16, 7p11.2, 18q, and 20q13.1 for physical inactivity, on 13q22-q31 and 18q for total physical activity, on 18 q for light and moderate activity, on 4q28.2, 7p11.2, 9q31.1 and 13q22-q31 for moderate and strenuous physical activity, and on 11p15 and 15q13.3 for time spent in physical activity [24, 25]. The 18q region, which has been linked to several physical activity phenotypes, flanks a candidate gene, MC4R, whose mutations have been associated with the development of obesity . MC4R is an important component of the regulation of energy balance, which is maintained by controlling energy intake, as food, and energy expenditure, as physical activity and metabolism. Association and animal studies with knockout and transgenic mice have confirmed the connection between MC4R and activity phenotypes and its involvement in the regulation of spontaneous and stimulus-induced activity [22, 27]. Other genes, such as gastrin-releasing peptide (GRP), striatin, transcripts encoding the endothelin B-receptor, and the cocaine- and amphetamine-regulated transcript (CART) have also been suggested as potential candidates to explain the variation of physical activity phenotypes [22, 24, 25].
The sedentarism, weekly PA, weekly PA_NS, and daily PA phenotypes were significantly influenced by additive genetic effects in the Baependi Heart Study population. The heritability estimates for physical activity (0.09-0.35) were lower than those found in twin studies (0.35 to 0.83) [18, 19, 29, 30], but they were within the range of estimates derived from family studies (0.06 to 0.66) [14–16, 20, 21, 25, 31]. The values for the heritability estimates in our study were lower for the weekly PA trait (0.19 and 0.26), and higher (0.23-0.60) for the sedentarism and daily PA traits than those reported in the literature [15, 16, 20, 21, 25, 32, 33]. In these studies, the experimental design varies in the methodology for parameter estimation, sample characteristics, and, mainly, in phenotype definition; therefore, a direct comparison of the heritability estimates is very difficult. In addition, only two of the studies were conducted with Mexican-American families [14, 20] that would have genetic backgrounds most similar to the high admixed Brazilian population; nevertheless, the phenotype acquisition and definition are distinct in relation to our study.
Genetic effects can be confounded by within-family transmission of behavioral patterns, thus household effects in the families of this population were investigated in all tested models. None of the analyzed phenotypes were significantly affected by a common household environment, consistent with a previous study ; however, the significance of the household effect on physical activity dimensions is inconsistent across previous studies. In the literature, shared environment components have been shown to have significant effects for both genders , in only one gender , in only one age group , or in specific phenotypes in a population . These inconsistencies are not surprising, as potential household effects may be heterogeneous and may represent unmeasured factors, such as diet, attitudes towards physical activity, family composition, neighborhood characteristics and cultural background. In addition, current residential addresses were used in our study to identify households and this could not capture long-term cumulative household effects, particularly those related to the formative years of the subjects .
Some studies have suggested that distinct genetic factors can influence the phenotypic expression of physical activity traits in males and females [29, 35], but this is still not well established. No evidence for a different set of physical activity genes in the two genders has been described in the literature [30, 36]. The heterogeneity in the results of the various studies indicates that knowledge about the genetic and environmental components of the physical activity variation is still incomplete. To assess this problem, we employed models in which heterogeneity in males and females was allowed in both variance components, in only the environmental variance component, or in only the genetic variance component. Significant gender differences were detected only with the models with no covariate effects. In the sedentarism phenotype, we observed heterogeneity in both variance components, with heritability estimates higher in females (0.22) than in males (0.05), and in the weekly PA trait, heterogeneity was seen in only the environmental variance component (heritability of 0.09 in females and 0.11 in males). There was no heterogeneity in variance components by gender for the weekly PA_NS trait. The models did not fit adequately for the daily PA trait, even when a robust estimation approach was applied, probably because the SOLAR package handled it as a quantitative trait. The heritability estimation was also performed for female and male data sets separately. Only the model with no covariate effects fitted to the female data. Using this model, the estimate in males was twice that for females. Considering the total phenotypic variance observed in males and females (1.241 and 0.435, respectively), we found that the level of daily physical activity in both genders was predominantly influenced by the environmental variance component, and this was higher in males than females. Similar [29, 35] and discordant [30, 36] results have been reported in the literature, suggesting that the heterogeneity in the variance components by gender is still not well defined for physical activity phenotypes.
Our study had some limitations. The first limitation was related to the questionnaire applied to assess the physical activity profile of this population. This questionnaire was based on the WHO-MONICA epidemiological instrument, but this has not yet been validated. Although our results are consistent with previous studies, they must be used with caution. Further, our study was based on self-reported answers to the questions and may suffer from phenotypic misclassifications in the data that could lead to underestimated heritability or to results with large standard errors; however, this is a common feature of other work on this topic. This issue could explain the low heritability estimates obtained for the sedentarism and daily PA phenotypes in our study. More objective devices such as accelerometers or pedometers have been proposed to collect physical activity data, but their application in large samples is difficult. The second limitation was the sample size. The data set used in this study is small in terms of subjects evaluated and controlled variables. The Baependi Heart Study is a longitudinal study and the first follow-up, which will permit a more effectively assessment of the physical activity profile of this population, has been started.
In conclusion, employing variance-components approaches, we have estimated the genetic influences on physical activity phenotypes in the Brazilian families that were part of the Baependi Heart Study. Heritability estimates for the sedentarism, weekly physical activity (weekly PA), weekly physical activity among non-sedentary subjects (weekly PA_NS), and level of daily physical activity (daily PA) traits were significant and from low to intermediate magnitude. Significant evidence for heterogeneity in variance components by gender was observed for the sedentarism and weekly PA phenotypes, but not for the weekly PA_NS trait. In the study population, the daily PA trait was predominantly influenced by environmental factors that had higher effects in males than in females. These data will add to the knowledge of the physical activity phenotypes in the Brazilian population and are concordant with the notion of significant biological determination in active behavior.
The authors thank the São Paulo Research Foundation (FAPESP-Grant 2007/58150-7) and the Hospital Samaritano, São Paulo for their financial support for this study. ARVRH acknowledges the São Paulo Research Foundation for its financial support (FAPESP-Grant 2010/51010-8). Some of the results reported in this paper were obtained using the SAGE software package that is supported by a U.S. Public Health Service Resource Grant (RR03655) from the National Center for Research Resources.
- Kesaniemi YK, Danforth E, Jensen MD, Kopelman PG, Lefebvre P, Reeder BA: Dose-response issues concerning physical activity and health: an evidence-based symposium. Med Sci Sports Exerc. 2001, 33 (6 Suppl): S351-358.PubMedGoogle Scholar
- WHO: Global recommendations on physical activity for health. 2010, Geneva: World Health OrganizationGoogle Scholar
- Brasil: Physical activity. Vigitel Brazil 2010: protection and risk factors for chronic diseases by telephone inquiry. 2010, Brasília: Ministry of Health, 152-Google Scholar
- Haskell WL, Lee IM, Pate RR, Powell KE, Blair SN, Franklin BA, Macera CA, Heath GW, Thompson PD, Bauman A: Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation. 2007, 116 (9): 1081-1093.View ArticlePubMedGoogle Scholar
- Lightfoot JT: Current understanding of the genetic basis for physical activity. J Nutr. 2011, 141 (3): 526-530. 10.3945/jn.110.127290.View ArticlePubMedGoogle Scholar
- Vink JM, Boomsma DI, Medland SE, de Moor MH, Stubbe JH, Cornes BK, Martin NG, Skytthea A, Kyvik KO, Rose RJ, et al: Variance components models for physical activity with age as modifier: a comparative twin study in seven countries. Twin Res Hum Genet. 2011, 14 (1): 25-34. 10.1375/twin.14.1.25.View ArticlePubMedGoogle Scholar
- Wolfarth B, Bray MS, Hagberg JM, Perusse L, Rauramaa R, Rivera MA, Roth SM, Rankinen T, Bouchard C: The human gene map for performance and health-related fitness phenotypes: the 2004 update. Med Sci Sports Exerc. 2005, 37 (6): 881-903.PubMedGoogle Scholar
- Oliveira CM, Pereira AC, Andrade M, Soler JM, Krieger JE: Heritability of cardiovascular risk factors in a Brazilian population: Baependi Heart Study. BMC Med Genet. 2008, 9: 32-View ArticlePubMedPubMed CentralGoogle Scholar
- S.A.G.E: Statistical Analysis for Genetic Epidemiology. 2009, 6.0.1Google Scholar
- Almasy L, Blangero J: Multipoint quantitative-trait linkage analysis in general pedigrees. Am J Hum Genet. 1998, 62 (5): 1198-1211. 10.1086/301844.View ArticlePubMedPubMed CentralGoogle Scholar
- Almasy L, Blangero J: Variance component methods for analysis of complex phenotypes. Cold Spring Harb Protoc. 2010, 2010 (5): pdb top77Google Scholar
- Giolo SR, Pereira AC, de Andrade M, Krieger JE, Soler JP: Evaluating gene by sex and age interactions on cardiovascular risk factors in Brazilian families. BMC Med Genet. 2010, 11: 132-10.1186/1471-2350-11-132.View ArticlePubMedPubMed CentralGoogle Scholar
- Brasil: Physical Activity. Vigitel Brazil 2006: protection and risk factors for chronic diseases by telephone inquiry. 2006, Brasília: Ministry of Health, 92-Google Scholar
- Mitchell BD, Rainwater DL, Hsueh WC, Kennedy AJ, Stern MP, Maccluer JW: Familial aggregation of nutrient intake and physical activity: results from the San Antonio Family Heart Study. Ann Epidemiol. 2003, 13 (2): 128-135. 10.1016/S1047-2797(02)00255-7.View ArticlePubMedGoogle Scholar
- Perusse L, Tremblay A, Leblanc C, Bouchard C: Genetic and environmental influences on level of habitual physical activity and exercise participation. Am J Epidemiol. 1989, 129 (5): 1012-1022.PubMedGoogle Scholar
- Seabra AF, Mendonca DM, Goring HH, Thomis MA, Maia JA: Genetic and environmental factors in familial clustering in physical activity. Eur J Epidemiol. 2008, 23 (3): 205-211. 10.1007/s10654-008-9222-x.View ArticlePubMedGoogle Scholar
- De Moor MH, Willemsen G, Rebollo-Mesa I, Stubbe JH, De Geus EJ, Boomsma DI: Exercise participation in adolescents and their parents: evidence for genetic and generation specific environmental effects. Behav Genet. 2011, 41 (2): 211-222. 10.1007/s10519-010-9415-4.View ArticlePubMedGoogle Scholar
- Joosen AM, Gielen M, Vlietinck R, Westerterp KR: Genetic analysis of physical activity in twins. Am J Clin Nutr. 2005, 82 (6): 1253-1259.PubMedGoogle Scholar
- Eriksson M, Rasmussen F, Tynelius P: Genetic factors in physical activity and the equal environment assumption-- the Swedish young male twins study. Behav Genet. 2006, 36 (2): 238-247. 10.1007/s10519-005-9018-7.View ArticlePubMedGoogle Scholar
- Butte NF, Cai G, Cole SA, Comuzzie AG: Viva la Familia Study: genetic and environmental contributions to childhood obesity and its comorbidities in the Hispanic population. Am J Clin Nutr. 2006, 84 (3): 646-654. quiz 673-644PubMedGoogle Scholar
- Choh AC, Demerath EW, Lee M, Williams KD, Towne B, Siervogel RM, Cole SA, Czerwinski SA: Genetic analysis of self-reported physical activity and adiposity: the Southwest Ohio Family Study. Public Health Nutr. 2009, 12 (8): 1052-1060. 10.1017/S1368980008003583.View ArticlePubMedGoogle Scholar
- Loos RJ, Rankinen T, Tremblay A, Perusse L, Chagnon Y, Bouchard C: Melanocortin-4 receptor gene and physical activity in the Quebec Family Study. Int J Obes (Lond). 2005, 29 (4): 420-428. 10.1038/sj.ijo.0802869.View ArticleGoogle Scholar
- Rankinen T, Bray MS, Hagberg JM, Perusse L, Roth SM, Wolfarth B, Bouchard C: The human gene map for performance and health-related fitness phenotypes: the 2005 update. Med Sci Sports Exerc. 2006, 38 (11): 1863-1888. 10.1249/01.mss.0000233789.01164.4f.View ArticlePubMedGoogle Scholar
- Simonen RL, Rankinen T, Perusse L, Rice T, Rao DC, Chagnon Y, Bouchard C: Genome-wide linkage scan for physical activity levels in the Quebec Family study. Med Sci Sports Exerc. 2003, 35 (8): 1355-1359. 10.1249/01.MSS.0000078937.22939.7E.View ArticlePubMedGoogle Scholar
- Cai G, Cole SA, Butte N, Bacino C, Diego V, Tan K, Goring HH, O'Rahilly S, Farooqi IS, Comuzzie AG: A quantitative trait locus on chromosome 18q for physical activity and dietary intake in Hispanic children. Obesity (Silver Spring). 2006, 14 (9): 1596-1604. 10.1038/oby.2006.184.View ArticleGoogle Scholar
- Zhou QY, Palmiter RD: Dopamine-deficient mice are severely hypoactive, adipsic, and aphagic. Cell. 1995, 83 (7): 1197-1209. 10.1016/0092-8674(95)90145-0.View ArticlePubMedGoogle Scholar
- Ste Marie L, Miura GI, Marsh DJ, Yagaloff K, Palmiter RD: A metabolic defect promotes obesity in mice lacking melanocortin-4 receptors. Proc Natl Acad Sci USA. 2000, 97 (22): 12339-12344. 10.1073/pnas.220409497.View ArticlePubMedPubMed CentralGoogle Scholar
- Farooqi IS, Keogh JM, Yeo GS, Lank EJ, Cheetham T, O'Rahilly S: Clinical spectrum of obesity and mutations in the melanocortin 4 receptor gene. N Engl J Med. 2003, 348 (12): 1085-1095. 10.1056/NEJMoa022050.View ArticlePubMedGoogle Scholar
- Maia JA, Thomis M, Beunen G: Genetic factors in physical activity levels: a twin study. Am J Prev Med. 2002, 23 (2 Suppl): 87-91.View ArticlePubMedGoogle Scholar
- Beunen G, Thomis M: Genetic determinants of sports participation and daily physical activity. Int J Obes Relat Metab Disord. 1999, 23 (Suppl 3): S55-63.View ArticlePubMedGoogle Scholar
- Simonen RL, Perusse L, Rankinen T, Rice T, Rao DC, Bouchard C: Familial aggregation of physical activity levels in the Quebec Family Study. Med Sci Sports Exerc. 2002, 34 (7): 1137-1142. 10.1097/00005768-200207000-00014.View ArticlePubMedGoogle Scholar
- Campos MAOA: Physical activity and metabolic syndrome components: a study in Azorean families. MsC thesis. 2006, Porto: University of PortoGoogle Scholar
- Campos MAOA: Combat the metabolic syndrome. Take care of your family and do physical activity. Porto: Regional direction of sports of the autonomous region of the Azores and FADEUP. 2007, 96-Google Scholar
- Simonen R, Levalahti E, Kaprio J, Videman T, Battie MC: Multivariate genetic analysis of lifetime exercise and environmental factors. Med Sci Sports Exerc. 2004, 36 (9): 1559-1566.PubMedGoogle Scholar
- Boomsma DI, van den Bree MB, Orlebeke JF, Molenaar PC: Resemblances of parents and twins in sports participation and heart rate. Behav Genet. 1989, 19 (1): 123-141. 10.1007/BF01065888.View ArticlePubMedGoogle Scholar
- Koopmans J: Smoking and sports participation. Genetic factors in coronary heart disease. Edited by: Goldbourt U. 1994, Dordrecht: Kluwer Academic Press, 217-235.View ArticleGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2350/12/155/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.