Statistical Genetics

About

The Quantitative Genetics Lab focuses on developing, enhancing, and applying novel statistical methods for a wide range of genetic studies, such as genome-wide association studies (GWAS) and twin and family-based heritability studies. Dr. Verhulst is the primary architect on the software package GW-SEM (Genome-Wide Structural Equation Modeling), which allows users to conduct structural equation modeling (SEM) on a genome-wide basis. The primary substantive area of research focuses on understanding the genetic factors that contribute to addictive behaviors. By looking across multiple addictive behaviors, within the context of various forms of comorbidity, it is possible to develop a more thorough understanding of addiction which will better inform prevention and treatment efforts.

The goal of the Statistical Genetics Lab is to develop and refine methods to analyze genetic data. This includes genetic studies that rely on genetic relatedness between family members, such as twin studies, as well as methods that utilize measured genetic such as genome-wide association approaches. The methodological advances from the statistical genetics laboratory focus primarily on Structural Equation Modeling techniques that can integrate seemingly disparate methods such as factor analysis or logistic regression into a unified framework.

Areas of Focus

Genome-Wide Structural Equation Modeling, or GW-SEM

The goal of GW-SEM is to provide users with the opportunity to analyze the complex, interconnected array of risk factors, biomarkers, environmental antecedents, comorbid disorders, and other health outcomes on a genome-wide basis using structural equation modeling techniques. We have published two drafts of the software. The first version (gwsem 1.0) provided proof of principle that SEM techniques could provide useful insights into complex trait genetics on a genome-wide scale (https://pubmed.ncbi.nlm.nih.gov/28299468/). The second version of the software (gwsem 2.0) has dramatically enhanced the flexibility, accessibility and efficiency of the algorithms and can easily be used to conduct multivariate genome-wide analyses (Pritikin et al. 2021).

Gene-by-Environment Interaction Methods

Gene-Environment Interactions (GxE) play a central role in the theoretical relationship between genetic factors and complex traits. Over the past decade, the Statistical Genetics Lab has published several manuscripts that explore the impact of GxE on a variety of different phenotypes, focusing on identifying best practices and avoiding pitfalls that can undermine the validity of the results. The current focus of GxE research within the Statistical Genetics Lab is to develop and extend statistical methods to conduct GxE models using genome-wide data to the fullest extent.

Anxiety and Internalizing Genetics

Anxiety and Internalizing genetics is one of the major phenotypic areas of interest for the Statistical Genetics Lab and is primarily conducted in conjunction with Dr. John ("Jack" Hettema. This interest has spanned multiple genetic models from genome-wide association studies (Hettema et al., 2019) to twin analyses (Savage et al., 2019), as well as the comorbidity between internalizing disorders and substance use behaviors (Moore et al., 2016).