For Academic Publishers
Recommendations for trusted, chemistry-friendly repositories
Journals should recommend trusted, chemistry-friendly research data repositories.
To assist authors in selecting well-established and community-specific repositories for their research data, trusted chemistry friendly repositories should be recommended by journals. These should be included within the author guidelines or the data policies.
Recommendations to include data availability statements
Journals should recommend that authors provide a data availability statement. Author guidelines should also provide templates for illustration.
Templates for data availability statements or a similarly termed section should be made available to authors in the journal's author guidelines. This guides authors in correctly submitting underlying data with their manuscript and effectively communicating how to find and access the data by including the DOI or other PID, enabling the FAIR principles (e.g. A1, I3). It should also be mentioned whether the templates are selectable and the information should be provided via the manuscript submission system or whether this information should be manually inserted into the initially submitted manuscript by the authors.
Data availability statements and manuscript submission systems
Journals should add a data availability statement to published articles and collect the necessary information through their submission systems.
Templates for data availability statements or a similarly termed section should also be added to the manuscript submission system. Once a template has been selected by the submitter, the data availability statement should be editable to allow authors to add additional information, such as what data are included in the dataset, similar to what is currently often mentioned in the section on supporting information PDF files. The submission system should then require the submitter to provide the necessary information, such as the DOI (specified as DOI name e.g. 10.1000/182 or as a URL i.e. including a resolver e.g. https://doi.org/10.1000/182 ), repository name, third party name and contact information, or reasons for restricted access and information on how to access a dataset, depending on the template used.
Link datasets to articles in CrossRef DOI metadata
Journals should use the information available in data availability statements to enhance CrossRef DOI metadata by linking articles to datasets.
With the DOI and repository name in hand, journals should enrich CrossRef DOI metadata of articles published following the FAIR principles (e.g F2, I3). This establishes a structured link between the DOI of the article and the DOI of the dataset and ensures humans and machines alike can interpret the relationship between the published objects. For CrossRef metadata, a related_item should be added to mention the name of the repository (equal to publisher in the corresponding dataset DataCite DOI metadata).
In XML:
<rel:program name="relations">
<rel:related_item>
<rel:description>
Dataset in <<<repository name>>>.
</rel:description>
<rel:inter_work_relation relationship-type="isSupplementedBy"
identifier-type="doi">10.prefix/suffix
</rel:inter_work_relation>
</rel:related_item>
</rel:program>
In agreement with Crossref's documentation on linking datasets to published items, the relationship type isSupplementedBy should be used.
Add data publications to prior publication policy
Journal author guidelines should explicitly encourage pre-submission of datasets.
Similar to the publication of preprints, journal author guidelines should allow for the pre-submission of datasets, as this is also already the case for many journals for crystallographic data published in CSD with CCDC. Datasets published prior to manuscript submission facilitate manuscript submission workflows as DOIs of datasets are registered. DOIs can therefore be validated and datasets can be included in the review process.
A disadvantage of pre-submission is that researchers cannot link the dataset to the manuscript, as the manuscript has not yet been published and no DOI has been registered. The metadata of the dataset must then be manually updated by the authors after the article has been published. Datasets with status under review are one way to overcome this disadvantage (see below).
Recommendations to include research data in the review process
Journal author guidelines should explicitly encourage research data to be included in the review process.
Some repositories have an under review status alongside the draft and published statuses. A dataset under review is not editable and not yet published, i.e. it does not have a DOI registered. Therefore, the DOI cannot be validated. Nonetheless, the dataset has an internally reserved DOI and is accessible via a URL to provide access to editors and reviewers. This allows research data to be included in the review process. The URL to access the dataset should be requested by the submission system so that it can be forwarded to editors and reviewers.
Encourage authors to publish datasets under review prior the articles gets published
Journal author guidelines should require that datasets with status under review to be published prior to the publication of the associated article.
To assist in automated workflows, such as linking the datasets to the published article through their respective PIDs, datasets under review should be published before the article gets published. Once a manuscript has been accepted, the authors should be informed to publish their dataset under review. This ensures that the data has a registered DOI when the article gets published. Consequently, journals can run quality control checks on the provided DOI such as validation. This process must be explicitly communicated with authors through the author guidelines, yet, can also be included within other communication upon acceptance. Contemporaneous, the DOI for the article should be provided so that authors can include this information in their dataset's metadata prior to the publication of the dataset. Finally, the article is published, and its DOI is registered.
Scholix.org
Journals and publishers should use Scholix.org.
Scholix provides the framework for improving the links between scientific literature and research data as well as between data and data with the goal of providing a high-level interoperability framework for exchanging information about these links. Thus, Scholix hubs, such as DataCite or OpenAire, contribute information on their metadata records, which contain information on connected digital objects. This information should be used by academic publishers to discover datasets that correspond to an article but were published after the article was published, which allows the metadata of the article to be updated with links to the dataset (see above).
Main authors: ORCID:0000-0003-4480-8661, ORCID: 0000-0002-6243-2840