Michael C. Neale - US grants

The etiology of many of our most serious social and medical problems, such as mental illness, cardiovascular disease and alcohol and drug abuse, appear very complex. Many currently funded research projects are collecting data from groups of relatives such as nuclear families, twin and adoptees in order to examine the role and importance of genetic factors in disease. Yet software for the analysis of these data, which often include longitudinal and multivariate components in additional to family structure, are inadequate. We aim to develop a comprehensive user-friendly resource for the statistical analysis of these data. Computational routines for matrix algebra, optimization and numerical integration will be interfaced to radically simplify specifications large-scale multivariate models and to enable specification of complex expressions for the likelihood of observed data. We will create a powerful menu-driven visual interface for modeling and analysis that will minimize user error and maximize flexibility. Data analysis will be streamlined throughout, from the reading of data, to local or remote computation, to the production of tables and figures of publication. This resource will be developed in preparation for collaborative data analysis on five currently funded projects spanning: cardiovascular disease, anxiety and depression, juvenile conduct disorder, drug and alcohol use and depression, and panic disorders, phobias, eating disorders, and schizophrenia. Use of the software will be taught at national and international workshops on methodology for twin and family studies, where further collaborative research initiatives will be identified. The code will be ported to a variety of platforms. A moderate level of support will be provided for users at other institutions worldwide.

DESCRIPTION (Adapted from applicant's abstract): This K02 Independent Scientist Award for Dr. Neale will enable him to expand and refocus his methodological efforts in psychiatric genetic epidemiology. From a background of developing software (the Mx package) to fit models to data from twins and families, the candidate will obtain further training in the areas of molecular genetics and clinical psychiatry. Skills in these areas will enhance the quality and increase the breadth of models that will be applied to datasets collected at MCV. These data include twin-family studies of adult female and male twins in Virginia, of schizophrenia in Ireland (PI Dr. K. Kendler), and of school-age twins in Virginia and North Carolina (PI Dr. L. Eaves). Specific foci for model development and application include: (I) models for comorbidity that use information on age at onset in relatives as well as familial comorbidity patterns to distinguish between alternative hypotheses; (ii) methods to control for variable age at onset in genetic studies of populations still at risk; (iii) environmental moderation of genetic risk; (iv) relapse and remission over time; (v) combination of information from multiple informants; (vi) the resolution of genetic heterogeneity and its relationship to psychiatric nosology; and (vi) the integration of data from genetic markers to partition genetic factors into those from specific quantitative trait loci and those from background polygenic variation. These models will be developed and tested with simulated data. They will be applied to data on generalized anxiety disorder, major depression, panic disorder, alcohol abuse/dependence, and schizophrenia in adults, and conduct disorder and attention deficit/hyperactivity disorder in the school-aged twins.

DESCRIPTION (provided by applicant): This application requests five further years of support for an Institutional National Research Service Award to cover multidisciplinary training in Psychiatric and Behavioral Genetics. We request support for three pre- doctoral and three post-doctoral students for primary training in: i) statistical, quantitative, behavioral and molecular genetics; ii) psychiatric nosology; iii) clinical or developmental psychology; iv) biostatistics. In addition to specializing in one of these primary areas, trainees will be encouraged to study at least one other area in sufficient detail to broaden their scope for future career development and interdisciplinary research. Training will usually be 4-5 years in duration for predoctoral and 2-3 years for postdoctoral students. Applications are expected from a broad array of disciplines, including medicine, psychiatry, psychology, biostatistics, neuroscience, molecular genetics and biology. Trainees are housed in the Virginia Institute for Psychiatric and Behavioral Genetics at Virginia Commonwealth University, a modern custom-built facility with private offices, state-of-the-art computational facilities and an integrated molecular genetics laboratory. Major strengths of the program include: i) broad expertise of faculty in areas of psychiatry, psychology, genetics and statistics; ii) highly productive research environment (faculty are in receipt of over 50 federally funded research grants, and are among the most highly cited in the field); iii) the extensive experience of the faculty in training at this level; iv) potential for trainees to take part in ongoing data collection and data analysis projects; v) access to large genetically informative datasets collected at VCU and other institutions in the past 25 years; and vi) access to a high throughput molecular biology laboratory and neuroimaging facilities.

[unreadable] DESCRIPTION (provided by applicant): The measurement of psychiatric symptoms and the subsequent diagnosis of disorders via DSM criteria may not be the optimal system for either clinical or research purposes. This project will develop and apply an array of methods to examine the empirical basis for the DSM nosological system and will develop more efficient methods of analysis of the available data. Virginia Institute for Psychiatric and Behavioral Genetics at Virginia Commonwealth University has assembled a large array of data collected from twins and other family members using state of the art interview schedules for the assessment of psychopathology. The datasets include the Virginia Twin Study of Adolescent Behavioral Development (VTSABD) -- a study of 8-16 year old twin pairs and their parents; the Adult Virginia Twin Studies (AVTS), and studies of adult twins and their parents, both of which contain longitudinal and multiple rater components. In addition, data from an epidemiologic study of juveniles, the Great Smokey Mountains Study (GSMS), and from adults, the National Comorbidity Survey (NCS) will be available for parallel analyses through collaboration with consultants. Existing methods that will be applied to these data include latent class analysis, item response theory, multiple rater models and modern regression methods. Extensions of these methods will be developed and applied, particularly multiple rater models for longitudinal data, multivariate contingent causal models for ordinal data, and structured versions of latent class and item response models that are suitable for the analysis of data collected from families or other clustered groups. Gender differences in symptom expression and subsequent development of clinical disorders will be assessed. Derived classes and trait scores will be validated in four main ways: covariance with risk factors; prediction of outcome over time; familial resemblance; and treatment response in clinical samples available at VCU, The long term goals of the project are to improve the efficiency of measurement, nosology, and knowledge of the etiology of common psychiatric disorders and their comorbidity.

This project aims to develop a series of novel approaches to phenotyping drug use and abuse. The general scheme is to develop statistical models from theory, implement them in user friendly software, and examine their statistical properties. Those models that perform sufficiently well will be applied to one or more sets of data to bring new insight into the assessment of substance use. The first goal is to extend of models for factorial invariance, which form the basis of testing for differences between groups. The primary extension will be to allow testing of invariance not merely between distinct groups, but also within groups that vary with respect to continuous variables such as age. This approach will be applied to confirmatory factor analysis, tc latent class analysis, and to models that represent mixtures of both factors and latent classes, and will be able to handle binary, ordinal and continuous observed variables. The method should prove valuable in assessing whether substance abuse patterns in the population represent continuous variation in liability or whether distinct groups exist. The second goal is to extend models for regime switching in the context of growth curve and other factor mixture models. This aim is intended to provide a better model for data that involve onset and offset of substance use, and to assist in uncovering heterogeneity. Third, we will develop methods for the analysis of certain forms of partially anonymized data such as those involving randomized response. These methods will be compared for their performance at detecting relationships with predictors, sequelae and correlates of partially randomized data, including resemblance between relatives and outcomes. All model development will be designed to permit the analysis and exploitation of data collected from relatives, and will include models for data on genetic markers, for both linkage and association studies. Applied data analyses will yield substantive results, guide model development, and test for robustness. An array of cross-sectional, longitudinal, and genetically informative datasets will be assembled and analyzed.

This application requests five further years of support for an Institutional National Research Service Award to cover multidisciplinary training in Psychiatric and Behavioral Genetics. We request support for three pre-doctoral students and four post-doctoral students for primary training in: (i) statistical/quantitative/behavioral/molecular genetics; (ii) psychiatric nosology; (iii) clinical or developmental psychology; (iv) biostatistics. In addition to specializing in one of these primary areas, trainees will be encouraged to study at least one other area in sufficient detail to broaden their scope for future career development, and to facilitate interdisciplinary research. Training will usually be 4-5 years in duration for pre-doctoral and 2-3 years for post-doctoral students, and will depend on the prior experience of the candidate. Applications are expected from a broad array of disciplines, including medicine, psychiatry, psychology, biostatistics, neuroscience, molecular genetics and biology. Trainees are housed in the Virginia Institute for Psychiatric and Behavioral Genetics at Virginia Commonwealth University, a modern facility with private offices, state-of-the-art computer technology and integrated molecular biology laboratory. Major strengths of the program include: (i) the broad expertise of the faculty in the areas of psychiatry, psychology, genetics and statistics; (ii) the highly productive research environment (faculty are currently in receipt of 35 federally funded research projects); (iii) the extensive experience of the faculty in training at this level; (iv) access to ongoing data collection and data analysis projects; (v) access to large, genetically informative datasets collected during the past 20 years; and (vi) access to high-throughput molecular biology laboratory.

DESCRIPTION (provided by applicant): This application seeks support from the National Institute on Drug Abuse R25 mechanism for pre-doctoral and postdoctoral research positions, infrastructure and workshops at the Virginia Institute for Psychiatric and Behavioral Genetics (VIPBG) at Virginia Commonwealth University (VCU). The overall goal of this research education program is to provide an environment for the innovative development and applications of statistical genetics methodology relevant to substance use, abuse and dependence (SUAD). Our research education program will be further developed with new courses in Statistical Innovation, Advanced Statistical Genetics, Epidemiology of Drug Abuse and Neuroscience to foster interdisciplinary research, a key component of NIDA's mission. The research education program consists of pre-doctoral and postdoctoral components. Pre-doctoral participants pursue degrees in Psychiatric, Behavioral and Statistical Genetics, Human & Molecular Genetics, Biostatistics or Computer Science. This component is designed to recruit potential future investigators to research in statistical genetics focused on SUAD. The aim is to create a cohort of PhD graduates who have been exposed to, and begun to publish research in, this area. The postdoctoral component recognizes that many promising researchers have training relevant to, but not focused on, the statistical genetics of SUAD. This flexible 2-3 year post-doctoral training component guides young investigators to this field of study and provides them with integrated training that enables them to pursue careers in statistical genetics of SUAD. Research education is intended for individuals with training in mathematics, statistics, biostatistics, genetics, psychology, computer science or pharmacology and to those who have completed their clinical requirements for the MD degree. The postdoctoral component aims to educate independent investigators who contribute to efforts to identify and characterize the genetic and environmental determinants of SUAD, its development, prevention and treatment. Accordingly we: i) Offer and carefully monitor a multidisciplinary integrated research training program with a wide range of research opportunities; ii) Meet the needs for training in emerging research areas in SUAD; iii) Provide formal education and intensive mentoring to researchers from diverse academic and ethnic backgrounds; iv) Offer a specialized curriculum that merges strengths in SUAD research at our institution with training in methodological innovation; and v) Disseminate course materials, developed software, user guides and example scripts to the wider community by teaching workshops and maintaining a website with webcasts, podcasts and script libraries.

Adolescence is a critical neurodevelopmental period associated with dramatic increases in rates of substance use. Identifying the pathways to substance use and its effects on adolescent development is critically important, as the effects of substance use during ongoing maturation likely have long-lasting effects on brain functioning and behavioral, health, and psychological outcomes. This Research Project Site application from the Twin Hub of the ABCD-USA Consortium is in response to RFA-DA-15-015; the proposal includes the University of Minnesota (hub leader), Virginia Commonwealth University, Washington University, and the University of Colorado to prospectively determine neurodevelopmental and behavioral predictors and consequences of substance use on children and adolescents. A representative community sample of 800 twin pairs, ages 9-10 years, from four sites whose researchers are leaders in twin research, SU and abuse, and neuroimaging of cognitive and emotional functioning will be tested, together with 700 singletons, contributing to the sample of 11,111 to be collected from 11 hubs across the ABCD-USA Consortium. Participants will undergo a comprehensive baseline assessment, including state-of-the-art brain imaging, comprehensive neuropsychological testing, bioassays, mobile monitoring and careful assessment of substance use, environment, psychopathological symptoms, and social functioning every 2 years. Interim annual interviews and quarterly web-based assessments will provide refined temporal resolution of behaviors, development, and life events with minimal participant burden. These Consortium-wide data obtained during the course of this project will elucidate: 1) effects of substance use patterns on the adolescent brain; 2) effects of substance use on behavioral and health outcomes; 3) bidirectional relationships between psychopathology and substance use patterns; 4) effects of individual genetic, behavioral, neurobiological, and environmental differences on risk profiles and substance use outcomes; and 5) ?gateway interactions? between use of different substances. The Twin Hub proposes to use classic and co-twin control designs to study genetic vs. environmental contributions to adolescent brain/behavioral development and how these contributions predict SU propensity. Using sophisticated growth trajectory modeling techniques, we will also identify the genetic and environmentally- determined consequences of substance use on brain and behavioral development, including the assessment of gene-by-environment interactions. In addition, we will develop biospecimen resources for future studies of genomic, epigenomic, metabolomic and microbiome changes that may influence substance use and its broad health consequences. Specific to this Twin-Hub, we will obtain baseline and follow-up serum, saliva, and in some cases gut microbiota from biological samples. This work enriches the full ABCD-USA Consortium given that disentangling G and E contributions to individual risk for addiction and sensitivity to SU's neurocognitive effects has highly significant public policy and prevention-based implications.

Abstract Neuroimaging has expanded our understanding of brain development from childhood into early adulthood. Adolescent substance use trends have shifted over time, but use of cannabis, alcohol, and tobacco remain prevalent, typically starting during teenage years, when serious mental health conditions also tend to emerge. Although physical health is at its lifetime peak, emerging concerns for teens include increasing rates of depression, anxiety, social isolation, suicidal ideation, and excessive use of screen media. The extent to which early substance use and other environmental exposures may place youth at risk for altered neurodevelopment and adverse outcomes remains poorly understood. A diverse sample of 11,878 9-10 year olds was enrolled from 21 sites across the ABCD Study consortium, and 554 were enrolled at Virginia Commonwealth University (VCU), under RFA-DA-15-015. All participants underwent a comprehensive baseline assessment, including state-of-the-art brain imaging, comprehensive neuropsychological testing, bioassays, careful assessment of substance use, mental health, physical health, culture and environment, and mobile monitoring every 2 years. Interim in-person annual interviews and biannual telephone or mobile app assessments provide refined temporal resolution of behaviors, development, and life events with minimal participant burden. Intensive efforts are made to retain the vast majority of participants through adolescence and beyond and retention rates thus far are very high. Data, securely and privately shared with the scientific community, will enable investigators to: (1) describe individual developmental trajectories in terms of neural, cognitive, emotional, and academic functioning, and influencing factors; (2) develop national standards of healthy brain development; (3) investigate the roles and interaction of genes and the environment on development; (4) examine how physical activity, sleep, screen time, sports injuries (including traumatic brain injuries), and other experiences affect brain development; (5) determine and replicate factors that influence the onset, course, and severity of mental illnesses; (6) characterize the relationship between mental health and substance use; and (7) specify how use of different substances (e.g., cannabis, alcohol, tobacco, caffeine) affects developmental outcomes, and how neural, cognitive, emotional, and environmental factors influence substance use risk.

Project Summary This project aims to develop and apply novel statistical approaches to address key causal questions in the of substance use, abuse and dependence. The development will focus on extending Mendelian Randomization methodology to new data types that test its key assumptions. These developments include: estimating and controlling for biases due to non- random mating; analyses of data from unrelated but genotyped individuals; extension to longitudinal data; multivariate network models; and multiple-group analyses to test for sex, age and other group differences. These new methods will be applied to unique longitudinal phenotype, genotype and methylation data on smokers and non- smokers from the Netherlands Twin Register, and data from the UK biobank. Precise identification of causal pathways will enable evidence- based prevention and treatment methods to be devised and implemented.

This application requests five further years of support for an Institutional National Research Service Award to cover multidisciplinary training in Psychiatric, Behavioral and Statistical Genetics. We request support for three pre-doctoral and three postdoctoral students for primary training in: i) statistical, quantitative, behavioral and molecular genetics; ii) psychiatric nosology; iii) neuroimaging genetics and neurobiology; iv) clinical psychology; iv) biostatistics. In addition to specializing in one of these areas, trainees will be exposed to all others and encouraged to study at least one other with sufficient detail to broaden their scope for future career development and interdisciplinary research. Training will usually be 4 years in duration for pre-doctoral and 2-3 years for postdoctoral students. Applications are expected to continue from a wide variety of disciplines, including medicine, psychiatry, psychology, biostatistics, neuroscience, molecular genetics and biology. Trainees are housed in the Virginia Institute for Psychiatric and Behavioral Genetics at Virginia Commonwealth University, a modern custom-built facility with private offices, state-of-the-art computational facilities, integrated molecular genetics and experimental laboratories, and an associated neuroimaging facility. Major strengths of the program include: i) broad expertise of faculty in psychiatry, psychology, genetics, neuroscience and statistics; ii) highly productive research environment with well-funded faculty who are among the most highly cited researchers in the field; iii) extensive experience and excellent track record of faculty in training at this level; iv) potential for trainees to take part in active data collection and data analysis projects; v) access to large genetically informative datasets collected at VCU and elsewhere; vi) direct access to genome sequencing, experimental study and neuroimaging facilities; vii) pairing with clinical psychiatrist to attend rounds; and viii) face-to-face instruction in responsible conduct of research from a leading author and instructor in this field.