Web-based Model Visualization / Creation for Structural Equation Models
PI: Tim Brick (Human Development and Family Studies)
Other Senior Team Members: Michael Hunter, Assistant Professor of Human Development and Family Studies, PSU; Michael Neale, Distinguished Professor of Psychiatry, Virginia Institute for Psychiatric and Behavioral Genetics; Steven Boker, Professor of Psychology and of Data Science, University of Virginia. The PI will serve as the mentor for the junior researcher.
The goal of this project is to develop an initial implementation of a user interface to specify complex models in the extended Structural Equation Modeling (xSEM) framework. Structural Equation Modeling is an approach to modeling common in the behavioral and health sciences whose primary value lies in the equivalence between an easily-understood network graph visualization of the model and the system of simultaneous equations defining the model’s expectations that can be numerically optimized. xSEM has become one of the most common approach for modeling covariance structures in behavioral science data; PubMed returns over 36 thousand papers using the phrase ‘Structural Equation Modeling’, with roughly 5000 papers a year in the past three years. However, modern tools for Structural Equation Modeling , such as the open-sourced OpenMx software for extended SEM (~1.6M total downloads & ~2000 citations), have recently expanded past the ability of these simple network visualizations to capture the complexity of the model. Although enhancements to the standards for SEM visualization have been proposed, few have been widely adopted, primarily due to a lack of software support.
This project will develop a web GUI for visualizing and building xSEM models with the ability to import and export model structures to the OpenMx statistical software for model fitting. Project goals will be: (1) develop a simple web tool for visualizing and creating simple SEM networks, (2) develop import and export capabilities for OpenMx models to allow the GUI to be used to run models using OpenMx, (3) as time allows, expand the capabilities of the tool to allow additional expansions of the xSEM framework, such as multi-level modeling capabilities, regularization, deep learning integration, and genome-wide complex trait analysis.
All direct statistical capabilities have been implemented in the back-end software, but there remain considerable challenges in the creation of sensible and usable visualizations and user experiences to understand the model, its fit to the data, and the relationships in the data that it implies.
A list of specific areas of computational and/or data science expertise or skills that the current team is particularly interested in recruiting to support the project: web development, statistical model visualization, data visualization. Experience with SEM or other covariance models helpful but not required.
Any other requirements or expectations of potential ICDS Junior Researchers: availability to meet regularly with the OpenMx team at their Friday 9-11 AM meeting.
A list of specific objectives for work supported by this call: open-source software GUI for structural equation modeling; possibly a conference or journal submission.
At least one medium to long-term goal: A collaborative proposal to NIH call for Building Sustainable Software Tools for Open Science (RFA-OD-24-010; submission due June 2026); at least one paper in a behavioral science methods journal describing the tool and its utility; contribution to (and authorship on) the OpenMx 3.0 release paper (expected in 2026). The previous two OpenMx papers are cited ~1000 times each.
A short statement (1 sentence to 1 paragraph) explaining the connection of the project to ICDS’s mission: The project directly advances the ICDS mission of advancing interdisciplinary research teams combining area expertise with advanced computational models. The OpenMx software this builds on is used in fields ranging across behavioral genetics, psychology/psychopathology, communications, health, addiction, mindfulness, education, child development, adolescent prevention, and aging.
A paragraph summarizing team member’s recent and/or planned engagement with ICDS: Timothy Brick is an ICDS co-hire and regularly participates in ICDS activities and committees, such as monthly lunches, faculty search committee, and annual ICDS symposia, and a faculty representative for ICDS and CHHD on the Chief Information Security Officer’s advisory board. He was recently appointed by the SSRI as a liaison with the ICDS for HPC development.