CIMeC pages on public repositories
This is a quick overview about CIMeC community pages on public repositories.
For any related question, please write to thinkopen.cimec@unitn.it or vittorio.iacovella@unitn.it specifying Data Sharing in the subject.
Introduction
Current practices in reporting require the researcher to prepare and share many other outputs beyond the "paper", such as experimental protocols, code, data, presentations, etc. This is often a task requested by funding agencies, together with the redaction of the Data Management Plan document.
Scientific journals may not be the preferred targets to host these outputs. There are several different options according to what you want to share. In order to facilitate this task, CIMeC manages community pages on several online archives.
Researchers could pick the target repository, share their contribution on their own and eventually, if they want, include it within the CIMeC community page.
According to the output to be shared, there could be several suggested choices:
| Contribution | Suggested choice |
|---|---|
| Presentation | Zenodo |
| Statement, Short Communication, Flyer | Zenodo |
| Single figure | Zenodo |
| Experimental protocol | OSF |
| Code | OSF, Github |
| FAIR Dataset | gin.g-node |
Check the following sections in order to pick the best repository according to your purposes.
CIMeC page on Zenodo
Zenodo is a Europe-based repository developed and organized by the European Commission through OpenAIRE to foster open circulation of knowledge.
Researchers can create a Zenodo account using the ORCID system, which is readily available through UniTrento login.
Uploading is performed by selecting Upload from the top right in the main page. Within the uploading form, there could also be picked a target community, including CIMeC UniTrento. Community will subsequently examine the request and eventually include the output in the community collection.
Uploading to the original author account occurs in any case. The record will earn a persistent Digital Object Identifier (DOI) which could be used for reporting purposes.
- Why Zenodo
One output, one landing page, one DOI: a single output can be uploaded on a specific page, no need to collapse many different outputs on the same reference. the page will also be linked to a persistent digital object identifier ready to be used in external references for reporting and altmetrics purposes. - Why not
If you need to upload datasets organized throughout a directory structure and you want other people to browse around to learn through directory hierarchy and names, an institutional repository (e.g. gin.g-node) might be a better choice.
CIMeC page on gin.g-node
The publication of structured collections of data can take place on the institutional repository gin.g-node, which is based on a version control and updating system for uploaded data, to facilitate both progressive updating and validation of uploaded data.
FAIR-ness
The repository responds to the four FAIR principles since it offers the possibility of obtaining a persistent Digital Object Identifier - DOI (Findable); data can be downloaded via web-browsing and / or command line (Accessible); the data must be loaded using a standard information representation scheme (Interoperable) and each dataset must be associated with a license that regulates its reuse (Reusable).
User Account
Setting a new account up can be made using the UniTrento institutional account through the “Register” option.
Uploading of a dataset is accomplished by creating a new repository and uploading the files.
Sharing
Once the dataset is completed, to make it appear on CIMeC Web page, please send an email to thinkopen.cimec@unitn.it with the subject “Dataset on the repository”.
Think Open account will fork the dataset (i.e.: create a snapshot of the current version) and from that point on it will appear on the CIMeC page as well.
gin.g-node features
gin.g-node is explicitly designed for sharing data coming from neuroscience experiments:
- It handles large amounts of files or large files in an easy way;
- data are arranged directories which could be directly browsed online (this is not always possible with other repositories);
- files are arranged in such a way that they can be downloaded individually
- it seamlessly integrates version control
Issues in data preparation
- Files must be arranged according to a proper schema. The easiest one for data coming from cognitive neuroscience experiments is BIDS - Brain Imaging Data Structure.
- All the personal information from the data MUST be individuated and carefully removed. Keep in mind that indirect identifiers of personal information (eg. addresses of the research center, information about researchers involved, etc.) could still be present even after a bulk removal of personal data. Please, double-check all the information you are going to share in order to avoid potential revealing of information.
- Most of the data are not re-identifiable. This includes most of the experiments performed at CIMeC which does not include structural data coming from neuroimaging (see also below). Thanks to the features of the BIDS schema, after BIDS-ification the dataset should be safe for sharing (see for example Eke, Damian, et al. "Pseudonymization of neuroimages and data protection: Increasing access to data while retaining scientific utility." Neuroimage: Reports 1.4 (2021): 100053).
- Part of the data may still contain residual information coming from the experiments (eg. structural data coming from neuroimaging). In this case, if it is necessary to publish this kind of information, there could be shared an empty file that serves as a placeholder for the location of potentially reidentifiable data. Specific requests of potential external users will be then examined accordingly.
Before sharing any kind of data, please go through your whole collection and check if are there any potential sources of personal data included. Do not hesitate to contact thinkopen.cimec@unitn.it or vittorio.iacovella@unitn.it for discussions and related questions.