Data Availability
NexOneMed requires a data availability statement on every research article. What the statement must say, which repositories to use, and the acceptable grounds for restricting access.
Every research article carries a data availability statement. Readers should be able to find the data behind the conclusions, and where they cannot, they should be told plainly why.
The requirement
Make the minimal dataset underlying the results available at the point of publication. The minimal dataset is whatever is needed to reproduce the reported findings: the values behind means, the data behind each figure, and the observations behind any statistical test.
Deposit it in a repository with a stable identifier, and cite it in the article with its DOI or accession number. A statement that data are “available from the authors on request” is accepted only where one of the restrictions below applies — on its own it is not sufficient, because such requests often go unanswered.
Where to deposit
| Kind of data | Where it should go |
|---|---|
| Nucleotide and protein sequences | GenBank, ENA or DDBJ |
| Gene expression | GEO or ArrayExpress |
| Proteomics | PRIDE or ProteomeXchange |
| Structures | Protein Data Bank |
| Clinical trial data | A recognised controlled-access repository |
| Anything else | A general repository such as Zenodo, Dryad, figshare or the Open Science Framework, or an institutional repository that issues DOIs |
Where a field-specific repository exists for your data type, use it.
Acceptable restrictions
Some data cannot be made public, and we do not pretend otherwise. Restrictions are acceptable where:
- release would identify participants and anonymisation cannot remove that risk;
- the ethics approval or participants’ consent does not permit public sharing;
- the data belong to a third party and are not the authors’ to release;
- release is barred by law, or would endanger an individual or a protected species or site.
In each case the statement must name the restriction, name the body that imposed it, and give a contact — an institutional data access committee or named office, not simply the corresponding author — through which a qualified researcher can apply. “Data are confidential” without a route of access is not acceptable.
Privacy first. Never deposit directly identifying information. Removing names is not anonymisation: dates of birth, small-area geography, rare conditions and unusual combinations of variables can all re-identify a participant.
Code and materials
Where results depend on custom code or scripts, deposit them with the data or in a public repository, with enough documentation to run them, and cite the version used. Novel materials and reagents should be available to other researchers on reasonable terms.
Example statements
Fully open: “All data underlying the findings are available in [repository] at [DOI].”
In the article: “All data underlying the findings are within the article and its supplementary files.”
Restricted: “The data contain potentially identifying patient information and cannot be shared publicly. Access requests are considered by [committee or office], which may be contacted at [address]. The authors did not receive special access privileges.”
How this is applied
The statement is checked before review and again before publication. Reviewers may ask to see the data to complete their assessment. If data promised in a published article turn out not to be available, we treat it under the corrections policy, which may extend to an expression of concern.