Persistence of changes in the genetic covariance matrix after a bottleneck

Abstract.— Genetic variance, phenotypic variance, and the genetic covariance matrix ( G ) can change as a result of genetic drift. These changes will persist over time to some extent and will continue if population size remains relatively small. Nine populations founded by a single pair of Drosophila melanogaster were measured for a series of six morphological characteristics for a large number of parent-offspring families at both the third generation after the bottlenecks and after 20 generations. From these data, the phenotypic variance, additive genetic variance, and G were estimated for each line at each generation. Phenotypic and genetic variances were highly correlated over time, so that the measurements made at the third generation were predictive of the state of the population 17 generations later. Genetic covariances were also somewhat stable over time; however, the G matrices of some lines changed significantly over the intervening generations. This change did not return the populations toward their original state before the population bottlenecks. We conclude that the genetic covariance matrix can change as a result of mild genetic drift over a short span of time.     

Relative efficiency of unequal versus equal cluster sizes in cluster randomized trials using generalized estimating equation models

There is growing interest in conducting cluster randomized trials (CRTs). For simplicity in sample size calculation, the cluster sizes are assumed to be identical across all clusters. However, equal cluster sizes are not guaranteed in practice. Therefore, the relative efficiency (RE) of unequal versus equal cluster sizes has been investigated when testing the treatment effect. One of the most important approaches to analyze a set of correlated data is the generalized estimating equation (GEE) proposed by Liang and Zeger, in which the “working correlation structure” is introduced and the association pattern depends on a vector of association parameters denoted by ρ. In this paper, we utilize GEE models to test the treatment effect in a two‐group comparison for continuous, binary, or count data in CRTs. The variances of the estimator of the treatment effect are derived for the different types of outcome. RE is defined as the ratio of variance of the estimator of the treatment effect for equal to unequal cluster sizes. We discuss a commonly used structure in CRTs—exchangeable, and derive the simpler formula of RE with continuous, binary, and count outcomes. Finally, REs are investigated for several scenarios of cluster size distributions through simulation studies. We propose an adjusted sample size due to efficiency loss. Additionally, we also propose an optimal sample size estimation based on the GEE models under a fixed budget for known and unknown association parameter (ρ) in the working correlation structure within the cluster.     

Estimation of covariance matrix of multivariate longitudinal data using modified Choleksky and hypersphere decompositions

Linear models are typically used to analyze multivariate longitudinal data. With these models, estimating the covariance matrix is not easy because the covariance matrix should account for complex correlated structures: the correlation between responses at each time point, the correlation within separate responses over time, and the cross-correlation between different responses at different times. In addition, the estimated covariance matrix should satisfy the positive definiteness condition, and it may be heteroscedastic. However, in practice, the structure of the covariance matrix is assumed to be homoscedastic and highly parsimonious, such as exchangeable or autoregressive with order one. These assumptions are too strong and result in inefficient estimates of the effects of covariates. Several studies have been conducted to solve these restrictions using modified Cholesky decomposition (MCD) and linear covariance models. However, modeling the correlation between responses at each time point is not easy because there is no natural ordering of the responses. In this paper, we use MCD and hypersphere decomposition to model the complex correlation structures for multivariate longitudinal data. We observe that the estimated covariance matrix using the decompositions is positive-definite and can be heteroscedastic and that it is also interpretable. The proposed methods are illustrated using data from a nonalcoholic fatty liver disease study.     

Model diagnostic tests for selecting informative correlation structure in correlated data

In the generalized method of moments approach to longitudinaldata analysis, unbiased estimating functions can be constructedto incorporate both the marginal mean and the correlation structureof the data. Increasing the number of parameters in the correlationstructure corresponds to increasing the number of estimatingfunctions. Thus, building a correlation model is equivalentto selecting estimating functions. This paper proposes a chi-squaredtest to choose informative unbiased estimating functions. Weshow that this methodology is useful for identifying which sourceof correlation it is important to incorporate when there aremultiple possible sources of correlation. This method can alsobe applied to determine the optimal working correlation forthe generalized estimating equation approach.     

Optimal design of longitudinal data analysis using generalized estimating equation models

Longitudinal studies are often applied in biomedical research and clinical trials to evaluate the treatment effect. The association pattern within the subject must be considered in both sample size calculation and the analysis. One of the most important approaches to analyze such a study is the generalized estimating equation (GEE) proposed by Liang and Zeger, in which “working correlation structure” is introduced and the association pattern within the subject depends on a vector of association parameters denoted by ρ. The explicit sample size formulas for two‐group comparison in linear and logistic regression models are obtained based on the GEE method by Liu and Liang. For cluster randomized trials (CRTs), researchers proposed the optimal sample sizes at both the cluster and individual level as a function of sampling costs and the intracluster correlation coefficient (ICC). In these approaches, the optimal sample sizes depend strongly on the ICC. However, the ICC is usually unknown for CRTs and multicenter trials. To overcome this shortcoming, Van Breukelen et al. consider a range of possible ICC values identified from literature reviews and present Maximin designs (MMDs) based on relative efficiency (RE) and efficiency under budget and cost constraints. In this paper, the optimal sample size and number of repeated measurements using GEE models with an exchangeable working correlation matrix is proposed under the considerations of fixed budget, where “optimal” refers to maximum power for a given sampling budget. The equations of sample size and number of repeated measurements for a known parameter value ρ are derived and a straightforward algorithm for unknown ρ is developed. Applications in practice are discussed. We also discuss the existence of the optimal design when an AR(1) working correlation matrix is assumed. Our proposed method can be extended under the scenarios when the true and working correlation matrix are different.     

Use of the score test as a goodness-of-fit measure of the covariance structure in genetic analysis of longitudinal data

Model selection is an essential issue in longitudinal data analysis since many different models have been proposed to fit the covariance structure. The likelihood criterion is commonly used and allows to compare the fit of alternative models. Its value does not reflect, however, the potential improvement that can still be reached in fitting the data unless a reference model with the actual covariance structure is available. The score test approach does not require the knowledge of a reference model, and the score statistic has a meaningful interpretation in itself as a goodness-of-fit measure. The aim of this paper was to show how the score statistic may be separated into the genetic and environmental parts, which is difficult with the likelihood criterion, and how it can be used to check parametric assumptions made on variance and correlation parameters. Selection of models for genetic analysis was applied to a dairy cattle example for milk production.     

Properties of components of covariance of inbred relatives and their estimates in a maize population

Summary Properties of three parameterizations, denoted as the C-model, D-model and Q-model, for covariances of inbred relatives under assumptions of no linkage or epistasis are explored and compared. Additive variance in an inbred population with inbreeding coefficient F, ² _AF =(1+F) ² _A where ² _A is additive variance in a panmictic population, if Q-model parameters Q _xx and Q _xy are both zero. Conditions sufficient for this to hold are presented in terms of gene frequencies and dominance contrasts (homozygotes vs. heterozygotes). Some other properties and potential uses of estimates of components in the models are also discussed. Estimates of components in the D-model and Q-model were calculated from a maize (Zea mays L.) study from which estimates of components in the C-model were previously published. Of particular interest were the covariance (Q _xy ) of effects of alleles at complete homozygosity with inbreeding depression effects, the covariance (D ₁) of additive effects at panmixia with inbreeding depression effects and the within-locus variance (D ₂, alias Q _xx ) of inbreeding depression effects. Estimates of Q _xy , D ₁, and D ₂ were small and nonsignificant in most cases. For ear height in the second year of the study, D ₂ appeared to be a major component. In some cases, results were obtained which had contradictory implications (negative D ₂ coupled with positive Q _xy or D ₁, and positive D ₂ coupled with negative ² _D ). A negative estimate of one or the other of ² _D or ² _A was obtained in one of the two within-year analyses for every character. Problems in getting realistic results were thought to be owing to excessive multicollinearity among the coefficients of the components in the expectations of the covariances of the kinds of relatives included in the study. Implications for future studies of this kind are discussed.Journal Article No. 87-3-14 of the Kentucky Agricultural Experiment Station published with the approval of the Director     

Generalized estimating equations: A pragmatic and flexible approach to the marginal GLM modelling of correlated data in the behavioural sciences

Within behavioural research, non‐normally distributed data with a complicated structure are common. For instance, data can represent repeated observations of quantities on the same individual. The regression analysis of such data is complicated both by the interdependency of the observations (response variables) and by their non‐normal distribution. Over the last decade, such data have been more and more frequently analysed using generalized mixed‐effect models. Some researchers invoke the heavy machinery of mixed‐effect modelling to obtain the desired population‐level (marginal) inference, which can be achieved by using simpler tools—namely by marginal models. This paper highlights marginal modelling (using generalized estimating equations [GEE]) as an alternative method. In various situations, GEE can be based on fewer assumptions and directly generate estimates (population‐level parameters) which are of immediate interest to the behavioural researcher (such as population means). Using four examples from behavioural research, we demonstrate the use, advantages, and limits of the GEE approach as implemented within the functions of the ‘geepack’ package in R.     

Empirical‐likelihood‐based criteria for model selection on marginal analysis of longitudinal data with dropout missingness

Longitudinal data are common in clinical trials and observational studies, where missing outcomes due to dropouts are always encountered. Under such context with the assumption of missing at random, the weighted generalized estimating equation (WGEE) approach is widely adopted for marginal analysis. Model selection on marginal mean regression is a crucial aspect of data analysis, and identifying an appropriate correlation structure for model fitting may also be of interest and importance. However, the existing information criteria for model selection in WGEE have limitations, such as separate criteria for the selection of marginal mean and correlation structures, unsatisfactory selection performance in small‐sample setups, and so forth. In particular, there are few studies to develop joint information criteria for selection of both marginal mean and correlation structures. In this work, by embedding empirical likelihood into the WGEE framework, we propose two innovative information criteria named a joint empirical Akaike information criterion and a joint empirical Bayesian information criterion, which can simultaneously select the variables for marginal mean regression and also correlation structure. Through extensive simulation studies, these empirical‐likelihood‐based criteria exhibit robustness, flexibility, and outperformance compared to the other criteria including the weighted quasi‐likelihood under the independence model criterion, the missing longitudinal information criterion, and the joint longitudinal information criterion. In addition, we provide a theoretical justification of our proposed criteria, and present two real data examples in practice for further illustration.     

Multilevel selection and the partitioning of covariance: a comparison of three approaches

Where the evolution of a trait is affected by selection at more than one hierarchical level, it is often useful to compare the magnitude of selection at each level by asking how much of the total evolutionary change is attributable to each level of selection. Three statistical partitioning techniques, each designed to answer this question, are compared, in relation to a simple multilevel selection model in which a trait's evolution is affected by both individual and group selection. None of the three techniques is wholly satisfactory: one implies that group selection can operate even if individual fitness is determined by individual phenotype alone, whereas the other two imply that group selection can operate even if there is no variance in group fitness. This has significant implications both for our understanding of what the term "multilevel selection" means and for the traditional concept of group selection.     

The evolution of sexual dimorphism in the house finch. I. Population divergence in morphological covariance structure

Abstract Patterns of genetic variation and covariation strongly affect the rate and direction of evolutionary change by limiting the amount and form of genetic variation available to natural selection. We studied evolution of morphological variance-covariance structure among seven populations of house finches (Carpodacus mexicanus) with a known phylogenetic history. We examined the relationship between within- and among-population covariance structure and, in particular, tested the concordance between hierarchical changes in morphological variance-covariance structure and phylogenetic history of this species. We found that among-population morphological divergence in either males or females did not follow the within-population covariance patterns. Hierarchical patterns of similarity in morphological covariance matrices were not congruent with a priori defined historical pattern of population divergence. Both of these results point to the lack of proportionality in morphological covariance structure of finch populations, suggesting that random drift alone is unlikely to account for observed divergence. Furthermore, drift alone cannot explain the sex differences in within- and among-population covariance patterns or sex-specific patterns of evolution of covariance structure. Our results suggest that extensive among-population variation in sexual dimorphism in morphological covariance structure was produced by population differences in local selection pressures acting on each sex.