Frequently asked questions Data Stewardship

You can find the guidelines and examples on this Radboudumc web page.

1. Use the RDM page example texts on Research data management – Radboudumc
2. Have your text reviewed by your supervisor
3. If possible: have your text reviewed by the data steward of your department
4. Send the text to datastewardship.im@radboudumc.nl for approval
5. Upload your text to HoraFinita

Yes, the same thesis guidelines apply.

Do not add your full DMP to your thesis. The DMP is a means to describe how RDM will be done and provides the exact storage locations. The RDM page is a summary of the DMP, including the DOIs of the published datasets, but excluding the exact locations of the archived data.

PhD students who started after January 2021 have to make a Data Management plan. It’s a good reference for the Research Data Management page in the thesis. Data Management Plans allow you to plan ahead, but you can change the plan whenever needed as it is a living document.

As of January 2021, Radboudumc researchers must adhere to F and A of FAIR according to the following guideline:

“For research data-based publications, of which one or more researchers from Radboud University/Radboudumc are authors, it must be clear how the associated data can be found (F) and the access management (A) of these data must be properly arranged.”

This means that data and metadata must be findable (F) by persons and computers outside Radboudumc, and it must be clear how access (A) to the data can be obtained. For data that cannot be traced back to individuals (anonymized data and animal test/lab data), the data must be openly accessible (open access). Data that is traceable to individuals (pseudonymized) should be published with 'restricted access'. In both cases a DOI (or other persistent identifier of link to the published dataset) should be included in the RDM paragraph.

To comply with this F and A guideline, the following has to be in place:

1. Research data that are suitable to be shared with other researchers (whether or not under certain restrictions or conditions) are archived in a certified and accessible research data repository. Metadata is always published in a repository, also when the data itself cannot be published. 
2. Generic metadata about the published dataset should be registered in the Research Information Services (RIS) interface of Radboud University. This is done automatically for datasets in the RDR. 

The RDR is Radboudumc's strongly recommended generic data repository. In the RDR, data can be stored during the study, data can be archived for the legal retention period and data can be shared and published for re-use.

For certain specific data types, such as proteomics, metabolomics or genetic data, publishing data in a discipline-specific repository is preferable because these data are better described and found there.

Datasets can also be shared and published in DANS data stations (e.g. Data Station Life, Health and Medical Sciences), which is an archive from KNAW or in Zenodo.

There are several other data repositories for different research disciplines:

• FAIRsharing: Searchable curated registry of databases, repositories, (inter-related to) data/metadata standards, and data policies by journals/publishers and funders.
• re3data.org: Registry of Research Data Repositories with detailed information about over 2000 research data repositories.

Data management SOPs are available in the Integral Quality System of the Radboudumc.

• SOP Data management
• SOP Management and Archiving of study documents

Pseudonymized data can be published under restricted access if the participants have given consent for sharing their data for reuse.

For publishing anonymous data, you do not need an informed consent: data that is anonymized should be published open access. Be aware that combination of items can also be identifying.

Data that cannot be pseudonymized or anonymized can only be made available for reuse through data visiting or controlled access. However, if you have relevant documentation that can be shared (e.g., scripts, aggregated data), this should still be published through a data sharing collection. Anonymized data can be archived in a data acquisition collection, which allows for sharing metadata in open access and assigns a DOI to the collection to make it possible to refer to the dataset. Data archived in a data acquisition is closed access and can only be accessed by collaborators who are granted access.

For WMO-compliant research, use the retention periods that are published on the CCMO website (Dutch website). For data collected in the context of studies not subject to the WMO (non-WMO research) a minimum retention period of 10 years is required by the Radboud University.

A persistent identifier is used to uniquely identify various objects.  Most data repositories provide a PID for published datasets. A Digital Object Identifier (DOI) is a type of PID.  All data collection types of the Radboud Data Repository receive a DOI. Also, other data repositories (e.g. DANS) use DOIs.

DANS has made a list of recommended sustainable file formats: File formats | DANS (knaw.nl)

The FAIR principles state that the data as such should be made findable, based on the metadata describing the dataset specifically. Data added to articles as supplementary materials are findable based on the metadata describing the article, but not the data per se. So, it is preferable to publish the data in a data repository, or at least refer to the supplementary materials link as part of the documentation of a published dataset.

The Radboudumc template is mandatory for:

DMP for WMO-compliant research and Local Feasibility (approval Radboudumc Executive Board) 
The Radboudumc DMP is one of the documents that needs to be reviewed during the Local Feasibility procedure. This is only applicable for WMO-compliant research for which the Radboudumc is the sponsor ('Verrichter') of the study.

DMP for PhD projects
The Radboudumc DMP template is mandatory for all PhD projects; a pdf export needs to be uploaded to Hora Finita three months after the start of the project, at the latest. If in doubt, for example because there is already an existing DMP for your research project, or because you are an external PhD student, contact RTC Data Stewardship.

DMP for funded research projects 
The Radboudumc template should also be used by researchers who receive a grant from NWO or ZonMw. ZonMw monitors the data management activities on the basis of a set of key items, which relate to several aspects of the FAIR principles. These key items must be fully delivered by the end of the project, and as much as possible at the start and during the project. The applicable items in the Radboudumc template are marked with **ZonMw key item**. If you are uncertain whether your funder allows the use of the Radboudumc DMP template, contact RTC Data Stewardship.

The Radboudumc DMP template is set up in a way that enables you to divide your projects into several parts, such as work packages, chapters or subprojects. This division can be maintained for all applicable questions in the plan. We advise combining as much as possible into one data management plan to avoid having to keep several plans up to date; the data management plan is after all a living document.

The Project Details and Write Plan are mandatory: provide as much information as possible in the Project Details, providing at least the project abstract and the project start and end date; the Write Plan section is the actual plan. The Contributors and Research Outputs sections are not mandatory because the plan itself (in ‘Write Plan’) contains all the relevant items from these sections.

Yes, all information about research data that will be collected, processed and/ or published should be included in the plan, which captures the whole research data life cycle. Publication of the data is an aspect, but not the only aspect, that is covered by a full data management plan.

You can find RadboudUMC specific information (e.g. how to request data collections, costs, etc) on the intranet.

General information about the Radboud Data Repository can be found on the RDR website.

The DRE offers an environment in which researchers can store, process and analyze their research data. It combines data storage and compute resources, whereas the RDR is only for storage. The RDR is used to archive raw research data and documentation, and to publish shareable data in accordance with the FAIR principles. Files that are stored in the RDR cannot be manipulated directly. If changes need to be made within a file, it needs to be changed locally first and then uploaded again. The RDR can be accessed from the DRE, allowing researchers to work with their (archived) raw data within the safe environment of the DRE and upload their documentation and processed data directly to the RDR without having to go through a difficult process of uploading and downloading to different systems.

In the RDR data collections must have at least 2 managers for continuity. One of the managers has to be data steward/data manager and the other is the main researcher. 

Preferably you choose a data steward/data manager of your department. If your department does not have a data steward, you can choose someone with a similar role. In case this is not applicable you can fill in the name of the PI.

There is no size limit to data collections. However, Radboudumc is allocated a certain amount of storage space based on (expected) use, so for large collections (> 1TB) we may have to acquire extra space, which means it may take a few days before the collection can be filled.

Raw (sensitive) data should be archived in a Data Acquisition Collection (DAC) of the Radboud Data Repository. The Data Sharing Collection is meant for publishing data. Please make sure that the identifying part of the data has been removed before adding data to the DSC. 

See also question How can we publish anonymized/pseudonymized data? in the Research Data Management FAQ section.

Yes. If it is not possible to share the whole dataset you can share a subset of data in a DSC. In the Radboud Data Repository the access level applies for the whole collection.Note that the files that you label as documentation files are treated as metadata: they will become downloadable to anyone once you publish your DSC.

In case you do not have access to your email or credentials you can access your RDR collections with an ORCID account. The other manager(s) of the collection can add you as a viewer or a contributor to the collection. Note that you cannot be added as a manager to the collection if you use ORCID login.

If you are added to the collection as a contributor, then you can make changes to the collection as long as the collection is not archived (DAC/RDC) or published (DSC).
After archiving a DAC and/or RDC it becomes read-only and is protected against changes, desired or otherwise. The collection manager can still update user access to archived collections, but they can no longer edit the content.

After publishing a DSC, the collection becomes read-only to ensure it is protected against any changes. Only a few metadata fields can still be changed by the collection manager.

Yes, each study has its own collections. The Data Acquisition Collection and Research Documentation Collection are meant for internal use and archiving. The Data Sharing Collection is meant for publishing data.

Radboudumc researchers can use the standard DUA of Radboudumc RUMC-RA-DUA-1.0 Template - Data Sharing Agreement (ru.nl).

Yes, this is possible if you provide a reason for asking for an embargo. You can publish a dataset in a Data Sharing Collection of the Radboud Data Repository with an embargo. During this period, no external parties (i.e. parties other than the added managers, contributors and viewers) can access your published collection's data. Only research administrators can place embargo periods on collections, so please contact your research administrator (datastewardship.im@radboudumc.nl) if you would like one or would like to edit the date.

The DRE offers an environment in which researchers can store, process and analyze their research data. It combines data storage and compute resources, whereas the RDR is only for storage. The RDR is used to archive raw research data and documentation, and to publish shareable data in accordance with the FAIR principles. Files that are stored in the RDR cannot be manipulated directly. If changes need to be made within a file, it needs to be changed locally first and then uploaded again. The RDR can be accessed from the DRE, allowing researchers to work with their (archived) raw data within the safe environment of the DRE and upload their documentation and processed data directly to the RDR without having to go through a difficult process of uploading and downloading to different systems.

Yes, anyone can request a myDRE account and be added to a workspace. It is the responsibility of the workspace Accountable and Privileged members to make sure access is revoked and/or a workspace is removed once the external collaborator no longer requires access to the data and the processing environment.

You can find more information on the costs of DRE on the intranet: DRE - Supported by RTC DS - Radboudumc

Visit our DRE information page for more information and the DRE support website for guides, requests and more.

RTC Data Stewardship offers a course on R basics for data preparation and presentation. You can read more and sign up for the course on Online Learning Environment.

The preferred environment for data processing and analysis is the Digital Research Environment (DRE). This environment allows for more freedom with regard to software, and also provides better compute power. Please first consider moving your work to the DRE.
For detailed instructions for improving the performance of R, see TOPdesk.

The instructions for troubleshooting R package installation in “Werkplek 2.0” can be found in TOPdesk.

Updating R on "Werkplek 2.0" isn't feasible due to the rapid release cycle of R. This can pose challenges when certain packages require a newer version of R. If you encounter such issues, consider the following solutions:

1. Alternative Packages: Look for other packages that can accomplish the same task. Often, R’s base functions can also provide a good alternative without needing an update.
2. Using DRE: Explore working with DRE where you have more flexibility with software updates and installations. This platform allows you to manage your own R version and package installations more freely.