Data Policy

1. Data type

ScienceDB is a public data repository dedicated to providing services for data sharing. At ScienceDB, we have no restrictions imposed on the discipline, subject matter, data size, or files format of submissions. It mainly shares and publishes the following data types: datasets, figures and tables of publications, slides, and code data.


Dataset refers to data organized in the form of file sets in any intelligible way regardless of structure, format, or size. At ScienceDB, we accept datasets from a full spectrum of disciplines across natural sciences, engineering technologies, and social sciences. Specifically, we consider data generated from basic research, applied research, experiments and assays, and raw data obtained through observation, monitoring, survey, investigation, measurement, testing, etc. and their derived data products.

Please note that dataset submissions should not contain manuscripts or published articles.

Figures and tables in publications

Figures and tables in publications specifically refer to those supplementary data documents pertaining to an unpublished manuscript or published article.


Slides refer to slides used for speeches, reports or presentations at academic conferences, which should contain proper references wherever appropriate.

Code data

Code data refers to program documentation such as source codes, pseudocodes for algorithm implementation, etc.

2. Data Community

Data Community is a form of data content organization and management under ScienceDB. It is managed by a specific Data Community Administrator. The Data Community relies on ScienceDB and ScienceDB’s operation & maintenance team provides related services and technical support for Data Community. ScienceDB encourages and supports multiple forms of data communities, such as journal data communities, institutional data communities, etc.

Each Data Community has its own dedicated Data Community Administrator. The Data Community Administrator formulates specific management regulations for the corresponding Data Community on the basis of ScienceDB’s Terms of Service and Data Policy. The Data Community Administrator is responsible for the data review in the corresponding Data Community under its own jurisdiction.

The Data Community and the Data Community Administrator are supervised by ScienceDB. Users can choose a specific Data Community to submit their data, and the data submitted by users will be reviewed by the Data Community Administrator of this Data Community, released on ScienceDB and included in this Data Community.

3. Codes of conduct for depositors

At ScienceDB, all the depositors are obliged to follow the codes of conduct below:

1)Promise to the data authenticity

All datas shall be produced in real scientific scenarios and shall no contain altered, fabricated, misleading or falsified contents.

2)Promise to the regulatory compliance

The data collection and generation process shall comply with regional and disciplinary laws and ethical codes, and the data contents shall not violate relevant regulations on open data sharing.

3)Statement on data rights

Data depositors and authors own the rights to publish, authorize, modify, utilize the data and to protect its integrity.

Data depositors shall confer on ScienceDB the rights to edit, copy, disseminate, translate, convert and print all or part of their work, in whatever means as specified or restricted by agreements.

Data depositors shall acknowledge all sponsorships of data production. Conflicts of interest shall also be noted at the depositors’ earliest convenience.

Data depositors shall confirm that the authorship list is organized in an order that reflects the authors’ contributions, and that all author information provided is correct, including authors’ affiliations, ORCID, etc. wherever applicable.

4)Policies concerning duplicate data publishing

Data depositors shall not publish a data on multiple data repositories, nor shall they submit a data simultaneously to multiple repositories for consideration.

In principle, data depositors shall not submit their published data to a second repository for review or publishing. Duplicate publishing is strictly prohibited except that data abstract can reappear as part of the authors’ speech or research article, or as an e-print at the authors’ discretion.

Depositors aiming for a data backup are obliged to provide the DOI of the published data, as well as specific citation information including author names, publisher, published time, version of the data, etc. . Depositors should also specifically mark their submission as a backup, and should ensure the data published under the same data license agreement as to its original version. At ScienceDB, we are not in a position to register DOI or CSTR for data backups.

If the submission is an update of a data previously published elsewhere other than ScienceDB, depositors are obliged to provide its original access link and describe the latest updates in their submission. Once accepted, it will be published under the same data license agreement as to the original. Under this scenario, we will be able to assign and register a DOI and a CSTR for the updated data.

5)Data with hazardous contents, data of animal or human research subject

Human-related research data

In the case of human-related research data, depositors are required to confirm by stating that they have obtained informed consent for the research, and that proper measures of data governance have been adjusted to protect the privacy of the human subjects. As a moral practice, data authors should also follow the relationship established with their informants.

Depositors are required to fill out and submit a Science Data Bank Data Desensitization Commitment Statement (for more in “Sensitive data masking ” section).

All submissions of research data involving human genetic resources shall conform to relevant administrative regulations. If the data is generated in China, depositors are required to include an approval letter on human genetic research (For more information, please refer to P. R. C Regulation on the Management of Human Genetic Resources,

Clinical trial data

All data submissions involving medical trials shall comply with Measures for the Ethical Review of Biomedical Research Involving Human (, and depositors are required submit a “trail registration number”. At ScienceDB, we only accept trial registration data issued by the registries from the WTO International Clinical Trials Registry Platform. Additionally, data depositors will be asked to sign and submit a Science Data Bank Data Desensitization Commitment Statement (for more in “Sensitive data masking” section). 

We do not consider clinical trial data without the documents as specified above.

Animal-related research data

For submissions of data involving animal experiments, authors shall conform to the ARRIVE guidelines or other equivalents, such as the UK’s Animals (Scientific Procedures) Act 1986, the EU’s Directive 2010/63/EU, the US’s Guide for the Care and Use of Laboratory Animals.

Data generated in China shall conform to the PRC Biosecurity Law, accessible at:

Authors should also indicate the gender of each animal and illustrate on its impact on their research findings.

Sensitive data masking

Data depositors are required to fill out and submit a Science Data Bank Data Desensitization Commitment Statement

In the statement, depositors shall state that all sensitive data have been effectively and irreversibly masked, and the methods of data masking can be described wherever necessary. Approval letters from the authors’ and depositors’ institution, if any, should be submitted together as an attachment.

It is authors’ and depositors’ responsibility to ensure no sensitive data in their submission. In case of privacy infringement or public security violation, authors shall agree to take all responsibilities, including mandatory withdrawal.

Submissions containing unmasked sensitive data will not be considered for publishing on ScienceDB.

Obligation to repot data errors

If, at any point, data authors and depositors become aware of major errors or inaccuracies in their published data, they shall contact ScienceDB for data update or retraction as soon as possible. They may also need to contact the editorial office of corresponding journals for paper revision or retraction whenever appropriate.

4. Data review criteria

This section specifies the review criteria we use to decide on a submission.

All data submissions shall meet ALL the criteria to be considered for publishing on ScienceDB. The criteria, as delineated below, apply to ScienceDB only and should not be taken as general rules or guidelines.

1)whether the submission meets our core metadata standards

For more information, please visit the Data Submission page or the Data Publishing Processes.

2)whether the metadata is consistent with the uploaded data files

We will check the consistency between metadata and data files uploaded, including their content, number of files, number of data entries, data volume, etc.

3)whether the data submitted is sufficiently intelligible

Submissions are considered as intelligible when, as a standalone entity, the data sufficiently describes the background and spatio-temporal span of data collection, the method of data generation, the device or computational model used, data accuracy, error range, etc.

Additionally, we check data usage notes, where authors are required to supplement information such as data dictionary and file naming convention to facilitate readers’ understanding and (re)use of the data.

We recommend depositors to provide information about data, including a description about column headings, units of measurements, abbreviations and what figures and/or analyses the data corresponding to.

Depositors are required to provide all access information (e. g. , DOI, CSTR) to facilitate the reader understanding whenever other academic sources (e. g. , data, GitHub projects, softwares, referenced articles) or associated resource entities (e. g. , derived data products) are used for data production.

Metadata should be a sufficient, clear and accurate description of the data.

We recommend depositors to use descriptive file names, giving data files names that are concise but indicative of their content.

4)whether the data submitted is complete.

We will check data integrity, including not only the integrity of data entries but also units of measurement. In case of incomplete or missing data, we ask depositors to briefly describe and explicate the case in Data Description.

5)whether the data submitted is accessible.

We will check whether any data files are damaged and whether they can be accessed in full. In whatever formats a data is stored, we ask all the files to be openable by the software application as specified by the and depositors in Data Description.

6)whether the submission contains sensitive contents.

We will check relevant documents or certificates when a data relates to a potential ethical issue.

7)whether the submission meets relevant codes of ethics.

We will check relevant documents or certificates when a data relates to a potential ethical issue.

8)whether the submission contains descriptions about the scientific value and reusability of the data.

In the Dataset description section, depositors are required to briefly describe the scientific value and reusability of their data submissions.

5. Data license agreements

ScienceDB provides seven data license agreements: CC0, CC-BY 4.0, CC BY-SA 4.0, CC BY-NC 4.0, CC BY-NC-SA 4.0, CC BY-ND 4.0 and CC BY-NC-ND 4.0.

For more information on license agreements, please refer to FAQ: What license agreements does ScienceDB support?

6. How to cite data published on ScienceDB

If any data published on ScienceDB contributes to your research, please cite it in the following format:

Author(s). Title. Data Version. Science Data Bank. DOI. (date published).
Author(s). Title. Data Version. Science Data Bank. CSTR No. (date published).

A specific example could be:

Wang Juanle, Chen Eryang, Zhu Junxiang, Zhou Yujie. Suspended solids concentration in the Poyang Lake 2000-2013 inversion of data collection. V1. Science Data Bank. (2015-07-15).

In addition, ScienceDB supports other data citation formats, such as GB/T 35294-2017 Information technology—Scientific data citation.

A data published on ScienceDB often correlates to a data paper. While it is not mandatory that users who cite the data also cite the data paper, we encourage our users to do so as an acknowledgment to the authors’ scholarly contribution.

7. Data retraction

All data published on ScienceDB are assigned a DOI and a CSTR, both of which are permanent identifiers. Published data can only be retracted in case of major data errors or academic misconduct, in which case only data files will be retracted while the metadata remains accessible.

For more information on data retraction, please refer to “FAQ: Can I retract a published data?"

8. Others

The overall data policies of ScienceDB abide by the Terms of Service of the site.

(last updated: January 18, 2022)