Fitting genetic models to twin data with binary and ordered categorical responses: A comparison of structural equation modelling and Bayesian hierarchical models

We compare Bayesian methodology utilizing free-ware BUGS (Bayesian Inference Using Gibbs Sampling) with the traditional structural equation modelling approach based on another free-ware package, Mx. Dichotomous and ordinal (three category) twin data were simulated according to different additive genetic and common environment models for phenotypic variation. Practical issues are discussed in using Gibbs sampling as implemented by BUGS to fit subject-specific Bayesian generalized linear models, where the components of variation may be estimated directly. The simulation study (based on 2000 twin pairs) indicated that there is a consistent advantage in using the Bayesian method to detect a correct model under certain specifications of additive genetics and common environmental effects. For binary data, both methods had difficulty in detecting the correct model when the additive genetic effect was low (between 10 and 20%) or of moderate range (between 20 and 40%). Furthermore, neither method could adequately detect a correct model that included a modest common environmental effect (20%) even when the additive genetic effect was large (50%). Power was significantly improved with ordinal data for most scenarios, except for the case of low heritability under a true ACE model. We illustrate and compare both methods using data from 1239 twin pairs over the age of 50 years, who were registered with the Australian National Health and Medical Research Council Twin Registry (ATR) and presented symptoms associated with osteoarthritis occurring in joints of the hand.

Designs for generalized linear models with several variables and model uncertainty

Standard factorial designs sometimes may be inadequate for experiments that aim to estimate a generalized linear model, for example, for describing a binary response in terms of several variables. A method is proposed for finding exact designs for such experiments that uses a criterion allowing for uncertainty in the link function, the linear predictor, or the model parameters, together with a design search. Designs are assessed and compared by simulation of the distribution of efficiencies relative to locally optimal designs over a space of possible models. Exact designs are investigated for two applications, and their advantages over factorial and central composite designs are demonstrated.

Optimal crossover designs for logistic regression models in pharmacodynamics

Pharmacodynamics (PD) is the study of the biochemical and physiological effects of drugs. The construction of optimal designs for dose-ranging trials with multiple periods is considered in this paper, where the outcome of the trial (the effect of the drug) is considered to be a binary response: the success or failure of a drug to bring about a particular change in the subject after a given amount of time. The carryover effect of each dose from one period to the next is assumed to be proportional to the direct effect. It is shown for a logistic regression model that the efficiency of optimal parallel (single-period) or crossover (two-period) design is substantially greater than a balanced design. The optimal designs are also shown to be robust to misspecification of the value of the parameters. Finally, the parallel and crossover designs are combined to provide the experimenter with greater flexibility.

Maximum Likelihood Estimation of Mixture Densities for Binned and Truncated Multivariate Data

Binning and truncation of data are common in data analysis and machine learning. This paper addresses the problem of fitting mixture densities to multivariate binned and truncated data. The EM approach proposed by McLachlan and Jones (Biometrics, 44: 2, 571-578, 1988) for the univariate case is generalized to multivariate measurements. The multivariate solution requires the evaluation of multidimensional integrals over each bin at each iteration of the EM procedure. Naive implementation of the procedure can lead to computationally inefficient results. To reduce the computational cost a number of straightforward numerical techniques are proposed. Results on simulated data indicate that the proposed methods can achieve significant computational gains with no loss in the accuracy of the final parameter estimates. Furthermore, experimental results suggest that with a sufficient number of bins and data points it is possible to estimate the true underlying density almost as well as if the data were not binned. The paper concludes with a brief description of an application of this approach to diagnosis of iron deficiency anemia, in the context of binned and truncated bivariate measurements of volume and hemoglobin concentration from an individual's red blood cells.

Class attainment among British men - A multivariate extension of the CASMIN model of intergenerational class mobility

The CASMIN Project is arguably the most influential contemporary study of class mobility in the world. However, CASMIN results with respect to weak vertical status effects on class mobility have been extensively criticized. Drawing on arguments about how to model vertical mobility, Hout and Hauser (1992) show that class mobility is strongly determined by vertical socioeconomic differences. This paper extends these arguments by estimating the CASMIN model while explicitly controlling for individual determinants of socioeconomic attainment. Using the 1972 Oxford Mobility Data and the 1979 and 1983 British Election Studies, the paper employs mixed legit models to show how individual socioeconomic factors and categorical differences between classes shape intergenerational mobility. The findings highlight the multidimensionality of class mobility and its irreducibility to vertical movement up and down a stratification hierarchy.

Estimating Two-Stage Models for Genetic Influences on Alcohol, Tobacco or Drug Use Initiation and Dependence Vulnerability in Twin and Family Data

Genetic research on risk of alcohol, tobacco or drug dependence must make allowance for the partial overlap of risk-factors for initiation of use, and risk-factors for dependence or other outcomes in users. Except in the extreme cases where genetic and environmental risk-factors for initiation and dependence overlap completely or are uncorrelated, there is no consensus about how best to estimate the magnitude of genetic or environmental correlations between Initiation and Dependence in twin and family data. We explore by computer simulation the biases to estimates of genetic and environmental parameters caused by model misspecification when Initiation can only be defined as a binary variable. For plausible simulated parameter values, the two-stage genetic models that we consider yield estimates of genetic and environmental variances for Dependence that, although biased, are not very discrepant from the true values. However, estimates of genetic (or environmental) correlations between Initiation and Dependence may be seriously biased, and may differ markedly under different two-stage models. Such estimates may have little credibility unless external data favor selection of one particular model. These problems can be avoided if Initiation can be assessed as a multiple-category variable (e.g. never versus early-onset versus later onset user), with at least two categories measurable in users at risk for dependence. Under these conditions, under certain distributional assumptions., recovery of simulated genetic and environmental correlations becomes possible, Illustrative application of the model to Australian twin data on smoking confirmed substantial heritability of smoking persistence (42%) with minimal overlap with genetic influences on initiation.

Vacancy solution theory for binary adsorption equilibria in heterogeneous carbon

A heterogeneous modified vacancy solution model of adsorption developed is evaluated. The new model considers the adsorption process through a mass-action law and is thermodynamically consistent, while maintaining the simplicity in calculation of multicomponent adsorption equilibria, as in the original vacancy solution theory. It incorporates the adsorbent heterogeneity through a pore-width-related potential energy, represented by Steele's 10-4-3 potential expression. The experimental data of various hydrocarbons, CO2 and SO2 on four different activated carbons - Ajax, Norit, Nuxit, and BPL - at multiple temperatures over a wide range of pressures were studied by the heterogeneous modified VST model to obtain the isotherm parameters and micropore-size distribution of carbons. The model successfully correlates the single-component adsorption equilibrium data for all compounds studied on various carbons. The fitting results for the vacancy occupancy parameter are consistent with the pressure change on different carbons, and the effect of pore heterogeneity is important in adsorption at elevated pressure. It predicts binary adsorption equilibria better than the IAST scheme, reflecting the significance of molecular size nonideality.

Bridging the gap between different statistical approaches: An integrated framework for modelling

This paper proposes a template for modelling complex datasets that integrates traditional statistical modelling approaches with more recent advances in statistics and modelling through an exploratory framework. Our approach builds on the well-known and long standing traditional idea of 'good practice in statistics' by establishing a comprehensive framework for modelling that focuses on exploration, prediction, interpretation and reliability assessment, a relatively new idea that allows individual assessment of predictions. The integrated framework we present comprises two stages. The first involves the use of exploratory methods to help visually understand the data and identify a parsimonious set of explanatory variables. The second encompasses a two step modelling process, where the use of non-parametric methods such as decision trees and generalized additive models are promoted to identify important variables and their modelling relationship with the response before a final predictive model is considered. We focus on fitting the predictive model using parametric, non-parametric and Bayesian approaches. This paper is motivated by a medical problem where interest focuses on developing a risk stratification system for morbidity of 1,710 cardiac patients given a suite of demographic, clinical and preoperative variables. Although the methods we use are applied specifically to this case study, these methods can be applied across any field, irrespective of the type of response.

Single-subject research designs and data analyses for assessing elite athletes' conditioning

Research in conditioning (all the processes of preparation for competition) has used group research designs, where multiple athletes are observed at one or more points in time. However, empirical reports of large inter-individual differences in response to conditioning regimens suggest that applied conditioning research would greatly benefit from single-subject research designs. Single-subject research designs allow us to find out the extent to which a specific conditioning regimen works for a specific athlete, as opposed to the average athlete, who is the focal point of group research designs. The aim of the following review is to outline the strategies and procedures of single-subject research as they pertain to.. the assessment of conditioning for individual athletes. The four main experimental designs in single-subject research are: the AB design, reversal (withdrawal) designs and their extensions, multiple baseline designs and alternating treatment designs. Visual and statistical analyses commonly used to analyse single-subject data, and advantages and limitations are discussed. Modelling of multivariate single-subject data using techniques such as dynamic factor analysis and structural equation modelling may identify individualised models of conditioning leading to better prediction of performance. Despite problems associated with data analyses in single-subject research (e.g. serial dependency), sports scientists should use single-subject research designs in applied conditioning research to understand how well an intervention (e.g. a training method) works and to predict performance for a particular athlete.

Multivariate monitoring of a biological wastewater treatment process: a case study at Melbourne Water's Western Treatment Plant

Biological wastewater treatment is a complex, multivariate process, in which a number of physical and biological processes occur simultaneously. In this study, principal component analysis (PCA) and parallel factor analysis (PARAFAC) were used to profile and characterise Lagoon 115E, a multistage biological lagoon treatment system at Melbourne Water's Western Treatment Plant (WTP) in Melbourne, Australia. In this study, the objective was to increase our understanding of the multivariate processes taking place in the lagoon. The data used in the study span a 7-year period during which samples were collected as often as weekly from the ponds of Lagoon 115E and subjected to analysis. The resulting database, involving 19 chemical and physical variables, was studied using the multivariate data analysis methods PCA and PARAFAC. With these methods, alterations in the state of the wastewater due to intrinsic and extrinsic factors could be discerned. The methods were effective in illustrating and visually representing the complex purification stages and cyclic changes occurring along the lagoon system. The two methods proved complementary, with each having its own beneficial features. (C) 2003 Elsevier B.V. All rights reserved.

Single patient (n-of-1) trials with binary treatment preference

The use of a fully parametric Bayesian method for analysing single patient trials based on the notion of treatment 'preference' is described. This Bayesian hierarchical modelling approach allows for full parameter uncertainty, use of prior information and the modelling of individual and patient sub-group structures. It provides updated probabilistic results for individual patients, and groups of patients with the same medical condition, as they are sequentially enrolled into individualized trials using the same medication alternatives. Two clinically interpretable criteria for determining a patient's response are detailed and illustrated using data from a previously published paper under two different prior information scenarios. Copyright (C) 2005 John Wiley & Sons, Ltd.

Normalized Gaussian Networks with Mixed Feature Data

With mixed feature data, problems are induced in modeling the gating network of normalized Gaussian (NG) networks as the assumption of multivariate Gaussian becomes invalid. In this paper, we propose an independence model to handle mixed feature data within the framework of NG networks. The method is illustrated using a real example of breast cancer data.

Multivariate genetic analysis of academic skills of the Queensland Core Skills Test and IQ highlight the importance of genetic g

This study examined the genetic and environmental relationships among 5 academic achievement skills of a standardized test of academic achievement, the Queensland Core Skills Test (QCST; Queensland Studies Authority, 2003a). QCST participants included 182 monozygotic pairs and 208 dizygotic pairs (mean 17 years +/- 0.4 standard deviation). IQ data were included in the analysis to correct for ascertainment bias. A genetic general factor explained virtually all genetic variance in the component academic skills scores, and accounted for 32% to 73% of their phenotypic variances. It also explained 56% and 42% of variation in Verbal IQ and Performance IQ respectively, suggesting that this factor is genetic g. Modest specific genetic effects were evident for achievement in mathematical problem solving and written expression. A single common factor adequately explained common environmental effects, which were also modest, and possibly due to assortative mating. The results suggest that general academic ability, derived from genetic influences and to a lesser extent common environmental influences, is the primary source of variation in component skills of the QCST.

Psychosocial and non-psychosocial risk factors for the new diagnosis of diabetes in elderly women

Objectives: Determine psychosocial variables associated with the new diagnosis of diabetes in elderly women. Examine whether variables remained significant predictors after controlling for non-psychosocial risk factors and the frequency of doctor visits. Research design and methods: A longitudinal cohort study was conducted using data from 10 300 women who completed a survey in 1996 and 1999. The women were aged between 70 and 74 years of age in 1996. The were asked to provide self-reports on a number of psychosocial and non-psychosocial variables in 1996 and on whether they had been diagnosed for the first time with diabetes in the 3-year period. The relationships between the potential risk factors and new diagnosis of diabetes were examined using binary logistic regression analysis. Results: Univariate results showed that not having a current partner, having low social support and having a mental health index score in the clinical range were all associated with higher risks of being diagnosed with diabetes for the first time. However the multivariate results showed that only a mental health index score in the clinical range and not having a current partner provided unique prediction of being newly diagnosed with diabetes. Of the non-psychosocial variables measured, only having a high BMI and hypertension were associated with increased risks of new diagnosis, while there was also evidence of a U shaped relationship between alcohol consumption and new diagnosis. Even after adjusting for frequency of doctor visits and non-psychosocial risk factors, a mental health index in the clinical range proved to still be a significant risk factor. Conclusions: A score on the mental health index that is within the clinical range is an independent risk factor for the new diagnosis of diabetes in elderly women. (c) 2006 Elsevier Ireland Ltd. All rights reserved.

Foetal malformations and seizure control: 52 months data of the Australian Pregnancy Registry

The Australian Pregnancy Registry, affiliated European Register of Antiepileptic drugs in Pregnancy (EURAP), recruits informed consenting women with epilepsy on treatment with antiepileptic drugs (AEDs), those untreated, and women on AEDs for other indications. Enrolment is considered prospective if it has occurred before presence or absence of major foetal malformations (FMs) are known, or retrospective, if they had occurred after the birth of infant or detection of major FM. Telephone Interviews are conducted to ascertain pregnancy outcome and collect data about seizures. To date 630 women have been enrolled, with 565 known pregnancy outcomes. Valproate (VPA) above 1100 mg/day was associated with a significantly higher incidence of FMs than other AEDs (P < 0.05). This was independent of other AED use or potentially confounding factors on multivariate analysis (OR = 7.3, P < 0.0001). Lamotrigine (LTG) monotherapy (n = 65), has so far been free of malformations. Although seizure control was not a primary outcome, we noted that more patients on LTG than on VPA required dose adjustments to control seizures. Data indicate an increased risk of FM in women taking VPA in doses > 1100 mg/day compared with other AEDs. The choice of AED for pregnant women with epilepsy requires assessment of balance of risks between teratogenicity and seizure control.