Comparison of analysis of variance and maximum likelihood based path analysis of twin data: Partitioning genetic and environmental sources of covariance

In order to investigate currently used model fitting strategies for twin data, analysis of variance (ANOVA) and path-maximum-likelihood (PATH-ML) methods of analyzing twin data were compared using simulation studies of 50 monozygotic (MZ) and 50 dizygotic (DZ) twin pairs. Phenotypic covariance was partitioned into additive genetic effects (A), environmental effects common to cotwins (C), and environmental variance unique to individuals (E). ANOVA and PATH-ML had identical power to detect total covariance. The PATH-ML AE model was much more powerful than ANOVA comparisons of rMZ and rDZ to detect A. However, to be unbiased, the AE model requires the assumption that C = 0.0. To allow use of the AE model to estimate A, the null hypothesis C = 0.0 is tested by comparing the goodness of fit of the ACE and AE models. Simulation of 50 MZ and 50 DZ pairs revealed that C must be greater than 55% of total variance before the null hypothesis would be rejected (P < 0.05) 80% of the time. Several recent publications were reviewed in which the null hypothesis C = 0.0 was accepted and apparently upwardly biased estimates of A, containing C, were presented with unrealistic P values. It was concluded that use of the AE model to estimate A gives an inflated view of the power of relatively small twin studies. It was recommended that ANOVA or comparison of the ACE and CE PATH-ML models be used to estimate and test the significance of A as neither requires that C = 0.0. © Wiley-Liss, Inc.     

Path analysis of seven fear factors in adult twin and sibling pairs and their parents

A multivariate path model of genetic and environmental transmission, used to derive expected covariances among adult twin and sibling pairs and their parents, was fitted to fear factor data from 250 twin families and 91 sibling families, with the use of a maximum-likelihood estimation procedure. The full model, with 259 free parameters, provides for parental cultural transmission, common twin and sibling environment, genotype-environment correlation, direct assortative mating, and genetic and environmental correlations. Significant effects were indicated for heritable transmission of common fears and phobias, and a single genetic factor accounted for most of the genetic covariance among the traits. Moderately high levels of common twin environment and a small effect for direct isomorphic marital assortment were also found. There was no evidence for cultural transmission, genotype-environment correlation, or common sibling environment.     

Genetic factors in the aetiology of mouth ulcers

The aetiology of mouth ulcers is uncertain, and prior research has indicated both environmental and genetic factors. In this study, information on mouth ulcer incidence was collected for 290 twin pairs—127 monozygous (MZ) and 163 dizygous (DZ)—and their parents, a total of 1,160 people. Self-reported data on mouth ulcer incidence were available for the twins, and in each family the mother also reported on the mouth ulcer incidence in the twins, the twins' father, and herself. A structural equation model—combining a measurement model, a rater bias model, and a model including genetic and environmental influences—was used to explain variation in mouth ulcers. The fitted model explained the variation in a latent phenotype of mouth ulcer incidence for the twins in terms of an additive genetic factor (64%), a common environment factor (26%), and a specific environment factor (10%). The mothers' ratings showed a significant positive bias. Genet. Epidemiol. 14:17–33,1997. © 1997 Wiley-Liss, Inc.     

Association between height and coronary heart disease mortality: a prospective study of 35,000 twin pairs

An inverse association between height and risk of coronary heart disease (CHD) is well demonstrated, but it is not known whether this association is because of genetic factors, socioeconomic background, or other environmental factors. Four population-based twin cohorts with register-based follow-up data on CHD mortality from Denmark (1966-1996), Finland (1975-2001), and Sweden (1963-2001 and 1972-2001) were used to investigate this question; response rates varied between 65% and 86%. Together, the cohorts included 74,704 twin individuals (35,042 complete twin pairs) with 5,943 CHD deaths during 1.99 million person-years of follow-up. Cox and conditional logistic regression models were used. Per 1-standard deviation decrease in height, height was inversely associated with CHD mortality in men (hazard ratio = 1.08, 95% confidence interval (CI): 1.04, 1.12) and in women (hazard ratio = 1.06, 95% CI: 1.01, 1.10). A twin who had died from CHD was on average shorter than the co-twin within monozygotic pairs (odds ratio = 1.27, 95% CI: 1.12, 1.44, with no sex difference), whereas a weaker association was found within dizygotic pairs in men (odds ratio = 1.01, 95% CI: 0.91, 1.13) and in women (odds ratio = 1.14, 95% CI: 1.01, 1.28). The inverse association between height and CHD mortality found within monozygotic discordant twin pairs suggests that this association is because of environmental factors that directly affect height and CHD risk.     

Genetic analysis of the age at menopause by using estimating equations and Bayesian random effects models

Multi-wave self-report data on age at menopause in 2182 female twin pairs (1355 monozygotic and 827 dizygotic pairs), were analysed to estimate the genetic, common and unique environmental contribution to variation in age at menopause. Two complementary approaches for analysing correlated time-to-onset twin data are considered: the generalized estimating equations (GEE) method in which one can estimate zygosity-specific dependence simultaneously with regression coefficients that describe the average population response to changing covariates; and a subject-specific Bayesian mixed model in which heterogeneity in regression parameters is explicitly modelled and the different components of variation may be estimated directly. The proportional hazards and Weibull models were utilized, as both produce natural frameworks for estimating relative risks while adjusting for simultaneous effects of other covariates. A simple Markov chain Monte Carlo method for covariate imputation of missing data was used and the actual implementation of the Bayesian model was based on Gibbs sampling using the freeware package BUGS.     

Multiple regression analysis of twin data obtained from selected samples

The multiple regression analysis of twin data in which a cotwin's score is predicted from that of a proband (the member of a twin pair selected because of a deviant score) and the coefficient of relationship provides a powerful test of genetic etiology (DeFries and Fulker: Behav Genet 15:467-473, 1985). Moreover, when an augmented model containing an interaction term is fitted to the same data set, direct estimates of heritability (h2) and the proportion of variance owing to shared environmental influences (c2) are also obtained. In the present paper, the expected partial regression coefficients estimated from these models are derived, and the flexibility of the general approach is illustrated. An extended model is formulated for the analysis of data from combined samples of affected and control twin pairs that yields tests for differential h2 and c2 in the two groups as well as pooled estimates of these parameters. The application of these models is illustrated by an analysis of data from reading-disabled and control twin pairs. Because of the ease, flexibility, and utility of the multiple regression analysis of twin data, it is an appealing alternative to more traditional model-fitting approaches.     

成年双生子血尿酸遗传度研究

目的 用双生子研究方法 对成年人血尿酸的遗传度进行估计.方法 从青岛双生子库募集成年双生子.测量身高、体重和血尿酸.相同性别的双生子采用16个多态标记进行卵型鉴定.通过校正年龄、性别和BMI,来构建结构方程模型估计遗传度.结果 共收集687对双生子数据,其中同卵双生子420对,异卵双生子267对.经平方根转换后,男性血尿酸水平(17.47±1.91)略高于女性(15.22±1.70)(P＜0.0001),通过校正年龄、性别和BMI后双生子血尿酸的组内相关系数分别为,同卵双生子0.70、异卵双生子0.40.运用性别限制模型进行拟合,最佳模型AE模型,加性别遗传因素和特殊环境因素共同作用血尿酸的水平.血尿酸的遗传度为70.5%(95%CI:65.9～74.6),特殊环境因素占29.5%(95%CI:25.4～34.2).结论 遗传因素是影响样本双生子血尿酸水平的主要因素.     

Sex-specific effects for body mass index in the new Norwegian twin panel

Sex-specific effects for body mass index (BMI) were explored in a newly established, population-based Norwegian twin panel. The sample includes 5,864 individuals, aged 18–25 years, who responded to a questionnaire containing items for zygosity classification, height, weight, health, health-related behaviors, well-being, and demographic information. Among the 2,570 intact pairs who returned the questionnaire there were 416 identical (MZ) male pairs, 387 fraternal (DZ) male pairs, 528 MZ female pairs, 443 DZ female pairs, and 796 unlike-sexed pairs. Alternate sets of models testing for either sex-specific genetic or environmental parameters were evaluated using structural equation analysis. Results from the most parsimonious model indicated that the genes contributing to variation in BMI are not identical for men and women; rather, some genetic effects were shared by the sexes and some were unique to each sex. Total variation in BMI could be explained by sex-specific additive genetic effects, as well as genetic and non-shared environmental effects common to men and women. Estimates of heritability were .708 for men and .789 for women, and the male-female genetic correlation was 0.622. The series of models specifying sex-specific shared environment also fit the data and suggests that shared environmental factors may be important for males but not for females. The findings raise questions concerning the relationship between sex-specific effects for BMI and sex differences in health outcomes. ©1995 Wiley-Liss, Inc.     

Genetic analysis of physical activity in twins

BACKGROUND: The reduced contribution of physical activity (PA) to daily energy expenditure contributes to the increased prevalence of obesity. A genetic control of activity-induced energy expenditure (AEE) may contribute to a genetic susceptibility to obesity. OBJECTIVE: Our aim was to investigate the relative contribution of genetic and environmental factors to the variation and covariation in AEE and PA. DESIGN: Twelve monozygotic and 8 same-sex dizygotic (including 2 same-sex sibling pairs; age differences: 2 and 2.5 y) twin pairs aged between 18 and 39 y participated. AEE was measured in a respiration chamber for 24 h and with doubly labeled water in daily life for 2 wk. PA was recorded simultaneously with a triaxial accelerometer. Structural equation modeling was used to separate and quantify the observed variance into sex-adjusted additive genetic and common and unique environmental contributions. RESULTS: In the respiration chamber, common and unique environmental factors explained the variance in AEE and PA, and no genetic contribution was found. In daily life, genetic factors explained 72% of the variance in AEE and 78% of the variance in PA. Unique environmental factors explained the remaining variance. The same additive genetic factors explained 67% of the covariance in AEE and PA in daily life. CONCLUSIONS: In the present exploratory study that used gold standard measurements for AEE and PA but a limited sample size, genetic influence explained a large part of the variation in AEE and PA in daily life, whereas both AEE and PA were influenced by environment only within the confined area of the respiration chamber.     

Heritability of Serum Apolipoprotein Concentrations in Middle-Aged Japanese Twins

Background

Studies of the genetic and environmental influences on apolipoproteins have been conducted, but few have used data from Japanese twins. The aim of this study was to quantify and compare the genetic and environmental causes of individual differences in the serum concentrations of apolipoproteins in Japanese middle-aged twins.

Methods

Apo A-I, apo A-II, apo B, apo C-II, apo C-III, and apo E were studied. A total of 142 twin pairs, aged 45 through 65 years, were enrolled: 85 monozygotic pairs (59 male, 26 female) and 57 same-sexed dizygotic pairs (43 male, 14 female). The intraclass correlation coefficient and structural equation modeling were used to estimate the best-fitting model and heritability.

Results

Sixteen percent to 75% of the total variances of apo A-I, apo C-II, and apo C-III were attributable to genetic influence; apo A-I and apo C-II were influenced by dominant genetic factors. Twenty percent to 73% of the total variances of apo A-II, apo B, and apo E were attributable to additive genetic influence; apo B was clearly influenced by common environmental factors. Furthermore, the heritability of all apolipoproteins was higher among females than among males.

Conclusions

Genetic factors, including additive genetic effects (A) and dominant effects (D), influence apolipoprotein levels. However, a common environment does not influence the variances of these apolipoproteins, with the exception of apo B. Furthermore, the heritability of apolipoprotein phenotypes differs by sex.Key words: apolipoprotein, heritability, adult twins     

儿童青少年双生子瘦素遗传度估计及影响因素分析

目的 分析影响儿童青少年瘦素的遗传因素和环境因素,探讨性别、年龄和体质量指数(BMI)的作用,为儿童肥胖早期预防提供依据.方法 选择6～18岁同性别双生子337对,平均年龄(12.3±3.5)岁,其中同卵双生子257对,异卵双生子80对.采用DNA微卫星多态性鉴定卵性.应用Mx结构方程模型分别计算年龄和BMI调整前后瘦素的遗传度,并检验性别、年龄和BMI对于模型的作用.结果 不同性别间身高、体重和瘦素水平差异均有统计学意义(P值均＜0.05).相关分析显示,瘦素水平与性别、年龄和BMI相关(P值均＜0.0l).遗传分析发现,调整前年龄方差在女生中影响较大,而男生则受共同环境方差影响较大.调整后男生特异性性别方差降低,最适模型为ACE(scale)模型.男、女生瘦素遗传模型一致,遗传度为20％.结论 儿童青少年人群中瘦素水平与性别、年龄和BMI相关.瘦素受遗传和环境因素共同影响.调整年龄及BMI后,瘦素遗传度不受性别影响.     

The heritability of cause-specific mortality: a correlated gamma-frailty model applied to mortality due to respiratory diseases in Danish twins born 1870-1930

The genetic influence on susceptibility to diseases of the respiratory system and all-cause mortality was studied using data for identical (MZ) and fraternal (DZ) twins. Data from the Danish Twin Register include 1344 MZ and 2411 DZ male twin pairs and 1470 MZ and 2730 DZ female twin pairs born between 1870 and 1930, where both individuals were alive on 1 011943. We used the correlated gamma-frailty model. Proportions of variance in frailty attributable to genetic and environmental factors were assessed using the structural equation model approach. For all-cause mortality the correlation coefficients of frailty for MZ twins tend to be higher than for DZ twins. For mortality with respect to respiratory diseases this effect was only seen in females, whereas males showed the opposite effect. Five standard biometric models are fitted to the data to evaluate the magnitude and nature of genetic and environmental factors on mortality. Using the best fitting biometric model heritability for cause of death was found to be 0.58 (0.07) for all-cause mortality (AE-model) and zero for diseases of the respiratory system for males. Heritability was 0.63 (0.11) for all-cause mortality (DE-model) and 0.18 (0.09) for diseases of the respiratory system (DE-model) for females. The analysis confirms the presence of a strong genetic influence on individual frailty associated with all-cause mortality. For respiratory diseases, no genetic influence was found in males and only weak genetic influence in females. The nature of genetic influences on frailty with respect to all-cause mortality is probably additive in males and dominant in females, whereas for frailty with respect to deaths caused by respiratory diseases in females, there are genetic factors present which are caused by dominance. Environmental influences are non-shared with exception of frailty with respect to respiratory diseases in males, where the shared environment plays an important role.     

Evidence for a strong genetic influence on childhood adiposity despite the force of the obesogenic environment

BACKGROUND: Body mass index (BMI) has been shown to be highly heritable, but most studies were carried out in cohorts born before the onset of the "obesity epidemic." OBJECTIVE: We aimed to quantify genetic and environmental influences on BMI and central adiposity in children growing up during a time of dramatic rises in pediatric obesity. DESIGN: We carried out twin analyses of BMI and waist circumference (WC) in a UK sample of 5092 twin pairs aged 8-11 y. Quantitative genetic model-fitting was used for the univariate analyses, and bivariate quantitative genetic model-fitting was used for the analysis of covariance between BMI and WC. RESULTS: Quantitative genetic model-fitting confirmed substantial heritability for BMI and WC (77% for both). Bivariate genetic analyses showed that, although the genetic influence on WC was largely common to BMI (60%), there was also a significant independent genetic effect (40%). For both BMI and WC, there was a very modest shared-environment effect, and the remaining environmental variance was unshared. CONCLUSIONS: Genetic influences on BMI and abdominal adiposity are high in children born since the onset of the pediatric obesity epidemic. Most of the genetic effect on abdominal adiposity is common to BMI, but 40% is attributable to independent genetic influences. Environmental effects are small and are divided approximately equally between shared and non-shared effects. Targeting the family may be vital for obesity prevention in the earliest years, but longer-term weight control will require a combination of individual engagement and society-wide efforts to modify the environment, especially for children at high genetic risk.     

Apolipoprotein E Genotype Frequency Patterns in Aged Danes as Revealed by Logistic Regression Models

Although the ApoE gene has been intensively studied in aging research, most of the studies conducted so far have been based on the traditional case-control design with subjects consisting of young controls and long-lived cases. The genotype frequency pattern in and between the two age-groups has been rarely investigated due to limitations in either research design or data analytical method. In this study, we genotyped 748 individuals (including both twin pairs and unrelated individuals) aged from 73 to 95 with aim at examining the genotype frequency trajectory of ApoE gene at high ages. Binomial and multinomial logistic regression models have been applied to model the gene frequency as a function of age and to investigate the modes of gene function (dominant, recessive, additive). The generalized estimation equations (GEEs) are introduced to account for the intra-pair genotype correlation in the twin pairs included in the data. Both the observed and the fitted frequencies show a constantly declining pattern of ApoE epsilon4 allele as age advances indicating a significant and steadily deleterious effect of the dominant allele that increases the hazard of death at high ages.     

The association between body height and coronary heart disease among Finnish twins and singletons

OBJECTIVE: An inverse association between body height and the incidence of coronary heart disease (CHD) has been observed. However, the mechanisms behind this association are still largely unknown. We will examine the role of genetic and familial factors behind the association in a large twin data set. DESIGN AND SETTING: The data were derived from the Finnish Twin cohort including 2438 singletons, 4073 monozygotic (MZ) twins, and 9202 dizygotic (DZ) twins aged 25-69 years at baseline in 1976. Incident CHD cases were derived from hospital discharge data and cause of death data between 1977 and 1995. Cox regression analysis and conditional logistic regression analysis were used. RESULTS: In population-level analyses no differences in the general risk of CHD between zygosity groups were found. The association between body height and CHD was similar between sexes and zygosity groups. When men and women in all zygosity groups were studied together an increased risk of CHD was found only among the shortest quartile (hazard ratio [HR] = 1.34, 95% CI: 1.14-1.57). Among the twin pairs discordant for CHD a suggestive increased risk for the shorter twin was seen among DZ twins (odds ratio [OR] = 1.19, 95% CI: 0.95-1.48) when men and women were studied together. CONCLUSION: An inverse association between body height and CHD was broadly similar between sexes and twin zygosity groups and was associated with short stature. Among discordant twin pairs we found a weak association among DZ twins but not MZ twins. This may suggest the role of genetic liability behind the association between body height and CHD.     

Elementary methods for the analysis of dichotomous outcomes in unselected samples of twins.

This paper presents an elementary statistical method for analyzing dichotomous outcomes in unselected samples of twin pairs using stratified estimators of the odds ratio. The methodology begins by first randomly designating one member of each twin pair as an "index" twin and the other member as the "co-twin." Stratifying on zygosity, odds ratios are used to measure the association between disease in the index twin and disease in the co-twin. From these zygosity-specific tables we calculate the Woolf-Haldane estimator of the common odds ratio (psi F, the weighted average of the zygosity-specific odds ratios), the Mantel-Haenszel test statistic (chi 2M-H) for the common odds ratio, and a test (chi 2G) for the difference in the zygosity-specific odds ratios. In this application, psi F provides an estimate of the familial association for disease and the accompanying chi 2M-H provides a test of the null hypothesis, psi F = 1 (i.e., there is no evidence for a familial influence on disease). The chi 2G is a test of the null hypothesis that psi MZ = psi DZ; a significant value for chi 2G suggests a genetic influence on disease (assuming that the observed odds ratios follow a pattern where psi MZ greater than psi DZ). A new test statistic (chi 2c) is proposed that incorporates the expectation that psi MZ = psi 2DZ under a purely additive genetic model with no common environmental effects. A significant value of chi c2 indicates that the different odds ratios across zygosity are partly due to common environmental influences. Conversely, a nonsignificant value of chi 2c is an indication that the zygosity-specific odds ratios are due solely to additive genetic effects and not to common environment. This basic approach is extended to examine the effects of measured indicators of the specific environment and the assessment of certain forms of gene by environment interaction. All of the methods are easily understood, highly flexible, readily computed using a hand calculator, and incorporate the inherent genetic information contained within twin samples.     

The Second to Fourth Digit Ratio (2D:4D) in a Japanese Twin Sample: Heritability, Prenatal Hormone Transfer, and Association with Sexual Orientation

The second to fourth digit ratio has been argued to reflect prenatal hormonal influences and is reportedly associated with various psychological and behavioral traits, such as sexual orientation, cognitive abilities, and personality. We examined genetic and environmental influences on the second to fourth digit ratio (2D:4D) using a Japanese twin sample (N = 300). The genetic analysis showed substantial additive genetic influences for both right and left hand 2D:4D. The rest of the variance was explained mainly by environmental influences not shared within twin pairs. These findings were, in general, in accordance with preceding studies with primarily Caucasian twin samples. The bivariate genetic analysis revealed that the additive genetic influences were largely shared between the right and left hand, while the non-shared environmental influences were largely unique to each hand. Results from a comparison of opposite-sex and same-sex twins were not significant, although they were in the predicted direction according to the prenatal hormone transfer hypothesis. Female monozygotic twin pairs discordant in sexual orientation showed significant within-pair differences in left hand 2D:4D, where non-heterosexual twins had lower (more masculinized) 2D:4D. In addition, we found that non-heterosexual male MZ twins had larger (more feminized) 2D:4D than their heterosexual co-twins. These results suggest the existence of non-shared environmental influences that affect both 2D:4D and sexual orientation.     

Genetic and environmental influences on binge eating in the absence of compensatory behaviors: a population-based twin study

OBJECTIVE: The current study explores the extent to which genetic and environmental factors influence liability to binge eating in the absence of compensatory behaviors (BE) in a population-based sample of twins. METHOD: Questionnaire data on 8,045 same-sex and opposite-sex twins, aged 18-31 years, from a Norwegian twin registry were used to assess BE during the last 6 months. RESULTS: The best-fitting biometrical model suggested that the heritability of BE was 41% (95% confidence interval [CI]: 0.31-0.50). Individual environmental factors accounted for the rest of the variance (59%; 95% CI: 0.50-0.69). No significant sex differences were found, but the statistical power to detect such effects was low. Shared environmental influences on the liability to BE in males could not be ruled out. DISCUSSION: The findings indicate significant additive genetic influences on BE, supporting the validity of the core features of binge eating disorder as a diagnostic category.     

人格障碍遗传度双生子研究

目的利用双生子人群估计人格障碍的遗传度.方法采用人格障碍诊断问卷第4版（PDQ-4）作为筛查工具,在知情同意的基础上,对青岛市20～70岁的成人双生子进行问卷调查.其中已知卵性鉴定结果的双生子共324对,对其中性别相同的242对双生子进行结构方程模型拟合,计算人格障碍的遗传度.结果人格障碍总评分遗传度为68.26%（60.26～74.78）,A组评分遗传度为59.00%（49.22～67.17）,B组评分遗传度为64.99%（56.24～72.16）,C组评分遗传度为63.66%（54.72～71.02）.分裂型、自恋型和依赖型有显著的遗传效应,遗传度分别为49.96%（37.94～60.14）、52.89%（41.85～62.24）和68.87%（60.80～75.40）.戏剧型不受遗传因素作用,共同环境效应显著,占总变异方差的54.08%（44.50～62.43）.结论人格障碍主要受遗传因素决定,这些结果将会对人格障碍的进一步研究提供依据.     

Effect of environmental and genetic factors on education-associated disparities in weight and weight gain: a study of Finnish adult twins

BACKGROUND: Disparities in body mass index (BMI) between persons with different educational levels in Western countries are well documented, but the background of these education-associated disparities remains poorly understood. OBJECTIVE: The objective was to examine the influence of environmental and genetic factors on education-associated disparities in self-reported BMI and weight change. DESIGN: Longitudinal postal surveys were performed in 1975, 1981, and 1990. The data were analyzed by using multivariate genetic models for twin data. The data derived from the Finnish Twin Cohort included 2482 monozygotic and 5113 dizygotic same-sex male and female twin pairs born between 1915 and 1957. RESULTS: Education-associated differences in BMI and in weight change were clear in 1975 and 1981, respectively, whereas no differences were seen in weight change between 1981 and 1990 when age and baseline BMI were adjusted for. The trait correlation between baseline BMI and educational attainment (-0.15 in men and women) was mainly due to correlations between additive genetic factors that contributed to BMI and education in men (-0.20; 95% CI: -0.25, -0.14) and women (-0.32; 95% CI: -0.40, -0.25) when adjusted for age. Among women, a weaker positive correlation was found for the unshared environmental effects contributing to the 2 traits (0.06; 95% CI: 0.02, 0.12). The same factors that affected the association between education and BMI in 1975 largely explained the association between education and weight change in 1981. CONCLUSION: The results suggest the possibility that common genetic factors affect educational attainment and body weight, which contribute to education-associated disparities in BMI in adulthood.