Repositories

A repository is a place where data and code can be stored or archived. There are different repositories for data, software, protocols, manuscripts, etc. It often includes metadata, containing details about the dataset (increasing findability).

Search for repository

If you are looking for a repository in a specific field or concerning a specific topic, you can search for repositories:

Re3data - search repositories

---

Make your data available

There are different ways to make your data available. You can for example make your dataset accesible for anyone by providing open access, or arrage for restricted/controlled access. If you are depositing data, you can contact the Data Access Committee for more information. The following repositories are well-used at our department for datasets:

If data is not of human origin:
European Nucleotide Archive

If data is of human origin (controlled access):
European Genome-Phenome Archive

---

Create a DOI

Always make sure to generate a persistent identifier (PID, for example: digital object identifier - DOI) for your dataset or code. Some repositories automatically generate a PID for you, others might not. You can generate a DOI for your code using Zenodo.

Close Repository Page

Data Management Plans

Drafting a data management plan is often een requirement when applying for grants, and it can help you in defining and structuring the requirements when handling your datasets.

A good Data Management Plan (DMP) contains the following aspects:

• - Description of the data that you will (re)use and generate during your project
• - Documentation and metadata of your data and code/scripts
• - Storage of your data, both during and after finishing your project
• - Policy about access to your datasets and data-sharing

The UBEC can provides assistance in setting up a Data Management Plan, contact us.

---

UMCU format DMP

Different requirements and rules on the format of a DMP apply, depending on the research council, funding agency or organisation where you’ve applied for funding. If no specific format is requested by the funder, please use the UMCU format (you can also download the PDF format here):
UMCU Data Management Plan

---

Other formats

If there is a specific format required, you can download it from the funders website. Below a list of available formats:
List of DMP formats

Close DMP Page

Licensing

When publishing your dataset or code, it is nessesary to select a suitable license. Below you can find a tools and the considerations that you need to be aware of when selecting your license.

There are multiple things you need to consider when selecting a licence for your dataset or for your code. You can use the license selector to go through the considerations step by step and select the correct license for your purpose.

If you are looking for the Creative Commons license selector, click here.

Close Licensing Page

FAIR data

In accordance to the FAIR principles both data and analysis should be Findable, Accessible, Interoperable and Reproducible.

Assistance with FAIR data

We provide input on how to make your research reproducible. One of the services we provide to achieve this is the FAIR check. Prior to publication you can provide us with your draft and we will check the following items;

• - Data availability (F and A)
• - Code availability (F and A)
• - Interoperability (I)
• - Reproducibility (R)

Based on our assessment we can provide your draft with a FAIR rating which can be included in the cover letter and final publication. If one of the components fails or can be improved you can work together with the UBEC to remedy this and increase your chances of publication.

We can also provide assistance on the complete FAIRification process. Contact us for more information.

---

Bring your own data workshop: FAIRifying datasets

Together with DTL and CGC, we've organized a 'Bring Your Own Data' workshop, intending to completely FAIRify the data of a research project. This workshop resulted in good proof-of-concept of the application of the FAIR principles, and the resulting deliverables can be used as a good example of what can be achieved when publishing data in a FAIR manner. Multi-omics datasets were used in this example and a FAIR data point (FDP) was deployed.

You can access the FAIR data point here.

For some examples of SPARQL queries that can be run on the FDP, click here.

The generated data models can be accessed here.

Close FAIR data Page

Data Paragraph

We provide a boilerplate text for grant applications that describes the services provided by UBEC/UMC Utrecht in the field of data management/stewardship.

The text provided can be used for the Data Management Plan paragraph that is present in your Grant-application. Please note that the information in this document assumes the default set-up of a project within the UMC Utrecht, Center for Molecular Medicine. If your project deviates from this standard, the default text should be edited to reflect those changes. Please contact us if this is the case, so we can include it in our repository and keep track of the possible scenarios and use cases.

The text is maintained and versioned at GitHub.

Close Data Paragraph Page

Data Access Committee

The UMCU is in the process of setting up a data access board to evaluate request for access to genomics data. Until the realisation of the centralised UMC Utrecht Data Access Board, the department specific ‘Division of Biomedical Genetics’ (DBG) Data Access Committee can be contacted to process and inform you of all matters regarding data sharing.

Researchers submitting data

• Submission statement (link to EGA website) (.docx)
• Data Access Requirements (.docx)

---

Researchers requesting data

• Data Request Form (.docx)
• Data Transfer Agreement (.docx)

---

Assessment of Data Request

When applying for data, the following form will be used for the assessment of your request:

• Data Request Assessment Form (.docx)

---

Data Access Committee Flow

The data access flow is visualised in the figure (click the figure for a larger version). In the top half of the figure, the application procedure is described (left for the UMC Utrecht researchers submitting data to a repository or publishing data, on the right for researchers applying for access to UMC Utrecht data). For the Devision of Biomedical Genetics, most of the datasets can be found in either ENA or EGA.
The lower part of the figure gives an indication of the application procedure.

For more information or assistance, please fill in the contact form on our website and the committee will contact you shortly.

Close DAC Page

Security and Privacy

There are certain measures you need to take into account regarding security and privacy.

As of the 25th of May 2018, the Algemene verordening gegevensbescherming (AVG) or General Data Protection Regulation (GDPR) is of effect. This means that on this date, the EU has one privacy law. You can find more information here.

---

Please take care of the following general security measures:

No personally identifiable information is allowed on portable, non-encrypted devices. Make sure to keep the personally identifiable information on UMCU servers. If you need to transfer data anyway, use an encrypted USB-drive or harddrive.

Do not use WhatsApp, Dropbox or sources alike for sensitive (patient) data. You can use Siilo if you want to securely app other health care professionals.

Always encrypt your harddrives, both portable and in your laptop.

Make sure to secure your data transfers. You can do this by using UMCU approved methods: encryption of your email (attachment) using the 'Veilig verzenden' button; use an SFTP server, use encrypted SURFdrive and always make sure to check the data integrity.

Use unique passwords for all your accounts. This means that you should never use the same passwords for multiple internet accounts. If one website is compromised, your other accounts will still be safe.

Set up two-factor authentication when possible.

Proper key management is important. You can encrypt your files, but if you do not handle your passphrases correctly, encryption won't matter. For example, make sure to send the password using another medium than that was used to transfer the files.

---

Data leaks

If you somehow stumble over personally identifiable information or accidentally caused a data leak yourself, please make sure to register this incident. Data leaks can result in very high fines (up until 820.000 euros), and it is advised to implement all above regulations.

---

ELSI helpdesk

If you need any help regarding ethical, legal and societal aspects or questions in your research project, you can contact the ELSI helpdesk. They can give you advice regarding for example informed consent and data protection. They are research oriented. It is of course also possible to contact the UMCU legal department with your questions.

Close Security and Privacy Page

Reproducibility

Below we've listed some tools and best practices that can help you to make your research more reproducible.

Protocols - Lab and bioinformatics

If you are working with wet lab protocols, you are advised to take a look at protocols.io. This platform can be used to do workflow management on your protocol in the lab. You can use it on your phone, where your protocol can be displayed on a 'step-by-step' basis. When you're working on a bioinformatics approach where you connect different tools and do manual steps in between, protocols.io can also be used to describe this workflow (see this link for an example of describing a methods section in a paper using this tool). You can use this tool to share your protocols, but it is also possible to create a 'private group'. You are advised to think about privacy and intellectual property concerns before using this tool.

---

Code: revision control systems

When you are working with code, you are advised to use a revision control system like Git or SVN. At the UMCU, there is a Institution level GitHub where you can park your code. In this group, it is also possible to make your repository private, if for some reason you do not wish to have outsiders looking at your code. If your code is even more sensitive, for example when there are intellectual property interests, we have a local GitLab repository that you can use. Please contact us using your contact form, and we will give you access.

When you are regularly using a revision control system, it is a good idea to think about the way you use it. There are multiple strategies available, read about the GitFlow strategy here.

If you are writing research software, you can consider joining the Netherlands Research Software Engineer community. They are trying to develop a strategy to make researchsoftware more reproducible.

If you are new to revision control systems, you can find a quick introductory course here.

---

Writing: revision control

When writing documents or papers in LaTeX, you can take a look at ShareLaTeX. This platform makes it possible to collaborate on a LaTeX document with mulitple users. The interface is also fairly intuitive, so users working mainly with Word, can easily contribute. Working with one collaborator is free, and there are paid plans available.

---

Suggestions?

If you have any tools/methods that you think should be added to the above list, please let us know!

Close Reproducibility Page

Better Science

Applying correct Data Stewardship can result in achieving better scientific results in the long run. It can result in saving money and lead to more exposure.

Michener et al. Nongeospatial metadata for the ecological sciences. 1997

More efficient workflow

A recent survey indicated that 60% of the time of a Data Scientist is spend on cleaning and organising data. This process would be more efficient if datasets are well described and interoperable. This means correct management and stewardship of datasets before publishing them.

A few methods that might improve the cleaning process:
- Structuring research projects
- Develop a Data Strategy (Data Management Plan)
- Workflow governance
- Correct use of Metadata
- Good documentation

---

Sharing: increasing citations and exposure

Publishing and sharing your datasets in a FAIR way, could lead to increased citations and exposure. Your analysis can be reviewed by others (ensuring consistency, reliability and repeatability) and new collaborations may arise. This does not only count for data, but also for your software/code/tools.

If you are correctly sharing your datasets, you might convice others to do the same. This will lead to a community with correctly described and interoperable datasets that can all be reused in other research project. This means that each research euro spend will result in more science.

As sharing becomes the standard, more datasets will become available for reuse by other scientists.

Close Better Science Page

Meet Legal Standards

One important aspect when doing (bio)medical research, is to comply to the law.

In the Netherlands there are several laws that are applicable to the research that we do at our department. The most important legislation is the AVG/GDPR.

---

AVG or GDPR

As of the 25th of May 2018, the Algemene verordening gegevensbescherming (AVG) or General Data Protection Regulation (GDPR) is of effect. This means that on this date, the EU has one privacy law. There are opportunities that can be recognized in this law:
- Data can move more freely across the EU
- Better control rights for patients and subjects
It also means extra, but not unnecessary, work. The AGV/GDPR states that more extensive registration is required to ensure privacy of the subjects. Who has access to the data and when? Is the data used for its intended purpose?

The 'Autoriteit Persoonsgegevens' has released a infographic about the GDPR.

You can find more information about the GDPR here. The full legislation is accessible here.

---

Sanctions

There are several sanctions if you are not compliant to the GDPR, ranging from warnings up until a fine of 20 million euro (or 4% of the annual turnover of the company, whichever number is greater).

Close Legal Page

Funders and Journals

More and more funders and journals are requiring correct application of data management and stewardship as a condition to funding/publishing.

As of 2017, Horizon 2020 project are required to use the FAIR Data Management Guidelines. This pilot aims to maximize access and re-use of data generated in research projects. There is a dedicated Data Management Plan developed to help researchers reach this goal.

ZonMw also requires researchers to a Data Management Plan.

It is expected that more funders and publishers will follow.

Close Funders and Journals Page