This story is part of a fall 2026 series celebrating 500 published datasets in the Duke Research Data Repository (RDR) by spotlighting 5 Duke researchers who make their work more transparent, discoverable, and reusable.

Joel Meyer is a Sally Kleberg Distinguished Professor in the Nicholas School of the Environment, Director of the Integrated Toxicology and Environmental Health Program, and the Associate Director of Undergraduate Studies. Dr. Meyer is also a Principal Investigator within the Superfund Research Center. The Superfund Research Center focuses on early, low-dose exposures to environmental contaminants and their developmental impacts, changes usually only evident later in life (NIEHS grant, P42ES010356). The Superfund Research Center has a collection within the Duke Research Data Repository to highlight the public data generated by the Center’s investigators and demonstrate compliance with NIH data management and sharing policy. The Collection currently contains almost 30 datasets; the Center also has shared data in disciplinary repositories linked on their website.

Dr. Meyer shared with us his thoughts on sharing environmental health data within the Duke Research Data Repository (RDR).

Tell us a little about the research and the kinds of data shared through the RDR. What should someone outside your field know about them?

My group studies the effects of pollutants on health, largely using cells in culture and the small nematode (worm) Caenorhabditis elegans. We gather a wide range of data, from molecular impacts (e.g., does the chemical cause DNA damage or oxidative stress) to cellular effects (e.g., does the chemical cause neurons to degenerate or malfunction) to whole-organism effects (e.g., does the chemical inhibit growth or reproduction). We often generate datasets with many thousands of data points per experiment.

Why do you choose to share your research data?

I think it is important that other researchers, as well as other interested people in general, be able to access the data for themselves. They might want to analyze it in a different way, or see if they agree with our analysis, or combine the data with other data for larger-scale studies. I also think it is our responsibility to make this data generally available since most of it was generated with taxpayer money. Finally, I hope that this also promotes a sense of scientific transparency.

What do you hope other researchers, or other audiences, might be able to do with the data you’ve shared?

I hope that it will be useful to regulators who are trying to protect the public from negative health effects of pollutants. I also hope that the data may be of use to community members, nonprofit organizations, and other researchers. In fact, we recently published a paper on how to do academic research in a way that is broadly useful (open access!), and the use of data repositories, like the RDR, is one of the things we highlight.

What has your experience been like working with the Research Data Repository team? Is there anything about the publishing or curation process that has been particularly useful to you?

It has been great! The team has been very fast to respond to questions and provide feedback, and very helpful regarding how to describe and structure the data for maximum utility. Frankly, they and the process have improved our own processes for organizing and storing data and metadata.

What would you say to another Duke researcher who is considering sharing their data?

Do it!

Explore Dr Meyer’s and other Duke researcher’s datasets published in the Superfund Research Center RDR Collection.

If your research grant or project would like to create a Collection within the RDR to more easily highlight datasets, feel free to contact us!

The RDR team thanks Dr. Meyer for his participation and supporting data publishing at Duke. For questions about the RDR contact: datamanagement@duke.edu

This story is part of a fall 2026 series celebrating 500 published datasets in the Duke Research Data Repository by spotlighting 5 Duke researchers who make their work more transparent, discoverable, and reusable.

Dr. Patrick Charbonneau maintains appointments in both Chemistry and Physics. Dr. Charbonneau published the 500th dataset within the Duke Research Data Repository (RDR) and is the top depositor. He shared with us his thoughts on data sharing and the use of the RDR.

Tell us a little about the research and the kinds of data shared through the RDR. What should someone outside your field know about them?

My group works at the intersection of theoretical chemistry and physics, using statistical mechanics and computational methods to understand complex systems. The datasets we share typically contain the numerical data behind published figures, along with code and documentation needed to understand or reproduce the results.

Why do you choose to share your research data? Has your thinking about data sharing changed over time?

Early in my career, archiving meant asking students and postdocs, usually as they were leaving, to organize their files on a departmental drive. I eventually realized that this approach was both stressful and fragile. We therefore shifted to preparing data, code, and documentation alongside each manuscript, so that sharing becomes part of publishing rather than something done retrospectively. An unexpected benefit is that preparing a dataset for others to use provides an additional check on the research itself.

What do you hope other researchers, or other audiences, might be able to do with the data you’ve shared?

I want our publications to remain reusable without having to track down a former student, postdoc, or me(!) years later. Surprisingly, this happened twice just this summer for papers from before we routinely deposited datasets; I have since deposited those data. But I also like that we cannot entirely predict what the data might eventually be useful for. A dataset created to answer one question may later become useful for comparison, methodological development, teaching, or a question we had never considered.

What has your experience been like working with the Duke Research Data Repository team? Is there anything about the publishing or curation process that has been particularly useful to you?

The RDR team has made data publication remarkably easy to incorporate into our research workflow. The curatorial oversight is particularly valuable: having another set of eyes on the organization, documentation, and long-term usability of a dataset helps turn files associated with a paper into a durable research product. Over time, working with the RDR has made depositing data simply part of what we consider completing a publication.

What would you say to another Duke researcher who is considering sharing their data?

Start small. There is no need to solve every possible question about what should be preserved before making a first deposit. Sharing even the numerical values behind published figures is a useful beginning, and once the process becomes part of the publication workflow, expanding what you share becomes much easier. In my experience, the cost is small, especially with the support of the RDR team, while the benefits for transparency, preservation, and future reuse can be substantial. Also, LLMs now make writing useful README files easier than ever.

Has anything interesting or unexpected happened as a result of sharing data?

In recent years, I started publishing articles on the history of science and related questions. Surprisingly, the datasets I deposited to accompany those articles—oral history interviews, textual corpora, and archival documents—are among the most consulted of all my deposits. I certainly didn’t expect my humanities datasets to be the popular ones!

Charbonneau’s humanities datasets: Data from: Sucre à la crème: Origin and Trajectory of an Authentic Québec Confection; Data from: Pralines des Voyageurs: An Iconic Intercultural Food; Data from: Elizabeth Monroe Boggs: From Quantum Chemistry to the Manhattan Project

Additional selected datasets:
Simons Collaboration on Cracking the Glass Problem RDR Collection
Data from: Decorrelation of the static and dynamic length scales in hard-sphere glass formers (500th dataset published in RDR)

The RDR team thanks Dr. Charbonneau for his participation and supporting data publishing at Duke. For questions about the RDR contact: datamanagement@duke.edu

We are happy to announce that the Duke Research Data Repository (RDR) has updated our platform to provide enhancements for data depositors. The platform is implemented in partnership with Duke University Libraries and TIND, a spinoff of CERN (see the press release). Below we explain what you need to know about this change.

We also have this short video demonstrating key changes and features.

What is changing?

• Homepage: There is a new landing page with slightly different navigation. All key documentation can be found under the “About” or “Resources” pages. You can still reach our site at research.repository.duke.edu.
• Links: While URLs have changed for datasets, all DOIs assigned to datasets will continue to work. We always encourage you to use DOIs (vs. URLs from your browser window) as those are the most stable and persistent!
• Persistent Identifiers (PIDs): All datasets will continue to receive DOIs, but we will also provide more robust support for ORCIDs (for unique identification of people) and RORs (for unique identification of organizations/institutions) for better tracking of research outputs and to comply with upcoming PID funder requirements
• Metadata: We will now integrate more DataCite metadata. Key metadata (descriptive information) will remain primarily the same; however, the form for describing your data will now have more built-in features for metadata standardization and compliance with best practices (see PIDs above).
• Dataset structure: All datasets now have a single landing page (no-sub pages as with the current platform). This resulted in certain datasets being slightly remodeled (e.g., folders zipped) to accommodate the new dataset structure. However, no files have been changed and data integrity, reuse, and reproducibility were prioritized in this process.

What do depositors need to know?

• File upload: For smaller datasets, you will now upload your files directly within the web form prior to hitting the “Submit” button (vs. via a Box link in the previous workflow). For larger datasets, upload will continue to be facilitated via Globus.
• Organization: Datasets will either need to be flat (no folders) or folders (if important to retain for access/reproducibility) will need to be packaged (e.g., zipped/tarred/etc) prior to upload.
• ORCIDs: To allow linking ORCIDs with datasets, we strongly encourage all dataset authors to have an ORCID prior to beginning a data submission. Go to the ORCID website to get your ORCID today!

What is new and improved?

• Submission dashboard: A new submission dashboard will allow depositors to track all current and past deposits in one easy location.
• Versioning: A new versioning module will allow depositors to more easily request the creation of a new version of their dataset. All previous versions will still be retained and the new version will receive a new DOI.
• File Previews: Previewing common file formats (e.g., tabular files, images, PDFs, zips) will now be available on the dataset landing page.
• Data Citations: You will now be able to copy data citations in a wide variety of citation styles.

What is staying the same?

• Curation by RDR staff: Datasets will continue to be reviewed by RDR staff prior to publication to help researchers make their data as FAIR (Findable, Accessible, Interoperable, and Reusable) as possible.
• Preservation and retention: Data will continue to comply with the DUL preservation policy including redundant copies, fixity checks, and over all information security as required by Duke. Our stated retention policy also will not change.
• Globus download: For datasets over a certain size threshold, users will continue to have the option to download those files via Globus (in addition to the option to download over the web browser). If the “Download from Globus” button appears at the top of a dataset, we encourage using Globus as downloading large scale data over a web browser has some challenges (e.g., timeouts, etc.).
• Embargoes: Depositors can continue to request embargoes for up to one year and we can continue to facilitate access to embargoed files for journal reviewers. Embargoed file names will now be viewable on the dataset page but cannot be downloaded until the embargo is lifted.
• Collections: Project-specific collections will still be supported and can be requested by emailing datamanagement@duke.edu.

The RDR curation team is excited to bring you this new and improved system and look forward to continuing to support data sharing, curation, and reproducibility for Duke generated data.

Please don’t hesitate to reach out with any questions at datamanagement@duke.edu.