Graduate students and postdoctoral researchers participated in a three-day computational toxicology workshop hosted by UL Research Institutes’ Chemical Insights and Duke University. Participants received hands-on training in open-source tools, publicly available chemical databases, new approach methodologies (NAMs), and chemical risk assessment techniques used to evaluate data-poor chemicals.
The future of research relies on dedication to open science. At UL Research Institutes’ Chemical Insights, we believe that to revolutionize science we must be dedicated to collaborating across fields, career levels, and sectors. As part of this dedication to openly sharing our knowledge, we hosted a three-day workshop, June 29 – July 1, 2026, at Duke University. About a dozen graduate and post-doctoral students received hands-on training in open-source tools and public databases used to assess data-poor chemicals.
Participants learned from Chemical Insights’ computational toxicology leaders who have helped advance the field throughout their careers:
Students mentioned that they appreciated the demos and how-to sessions coming from the professionals who helped create these databases. They also valued insights about which analyses and tools are commonly used in the field of computational toxicology. Multiple students mentioned that the insights gave them a good idea of how to incorporate models into their dissertation and future work.
“I think new scientists can be very creative. They bring new perspectives and creative ways to solve old problems. Making them aware of the tools that exist allows them to focus on the issues that need the most help.” – John Wambaugh, Ph.D.
Participants learned how to evaluate hazards, model exposures, interpret biomonitoring data, apply new approach methodologies (NAMs), and integrate diverse data streams to support chemical risk assessment and prioritization decisions.
Students participated in a hands-on exercise stepping through how to identify risk-relevant information for a list of chemicals using RMarkdown. They left the workshop with the script to perform the analysis and gain experience in understanding each step.
Throughout the workshop, students asked about how they can get their data into a public and computationally accessible format, highlighting that they are already thinking about how people can use the data they generate. One student asked about how to identify new chemicals or classes of chemicals to screen in high-throughput screening assays; for example, can we find the next per-and polyfluoroalkyl substances (PFAS) that are already in the human environment? This insightful question emphasizes the power of making data publicly available from different scientific domains, including chemistry, hazard, exposure, toxicokinetics, to identify the data-poor areas of chemical space.
As part of this new wave of collaborative thinking and dedications to open science, Chemical Insights is driven to put chemical data into the hands of others asking different questions.
Highlighting the importance of our ongoing research at Chemical Insights, students asked questions about what affects confidence in determining the bioactive concentration of a chemical: how many assays or which assays are needed to establish a “safe” concentration threshold? How does the distribution of chemicals in an assay will affect our ability to estimate the bioactive concentration?
This workshop wouldn’t be possible without a collaboration between Duke University and Chemical Insights leaders. We are thankful to the participants for attending and to those who helped make this workshop possible.