How to prepare and submit a manuscript to Scientific Data. Please see our Aims and scopefor advice on what to consider before doing this.

Key steps:

 

1. Select an article type

2. Deposit the dataset

3. Prepare the manuscript

   • General
   • Title & abstract
   • Authors & affiliations
   • Background & Summary / Introduction
   • Methods
   • Data Record
   • Data Overview (optional)
   • Technical Validation
   • Usage Notes
   • Data Availability
   • Code Availability
   • Acknowledgements, Author Contributions, Funding & Competing Interests
   • References
   • Citing Data
   • Figures & Tables
   • File requirements for first round review
   • Covering letters

4. Submit the manuscript

5. First round peer review

6. Revised manuscripts

   • What to expect
   • General requirements
   • Figures
   • Tables
   • Equations
   • Supplementary information
   • LaTeX
   • Consortia

1. Select an article type

 

Read more about these formats in our Aims & Scope section.

2. Deposit the dataset

 

Data Descriptors are associated with datasets that we review alongside the paper. The requirements for sharing depend on the peer review stage:

For first round review we require that data are available for download and review by reviewers via any URL that allows anonymous download. This can be the final, intended repository link if authors wish to deposit the data from the start, or a temporary location such as cloud storage if you wish to delay or embargo.

From the second round of review onwards we require that data are public and available in a formal data repository that meets our policy, in preparation for any final publication.

In most cases we recommend these are not separate steps, meaning the data is live at the repository from the start, as the simplest option. Please see our data repositories guidance for a full list of requirements, and some suggestions

Sharing is also required for Articles if relevant to the results of the work.

3. Prepare the&nbspmanuscript

 

General

Submissions should be clearly written and understandable by scientists from diverse backgrounds. Technical jargon should be avoided where possible and clearly explained where its use is necessary. Titles and abstracts should be written in language understandable to any scientist. Abbreviations should also be kept to a minimum and, where unavoidable, should be defined in the text at their first occurrence (please do not use abbreviations sections)

We do not have rules for format, font size, style, etc as long as the mandated headings for the article type are included. Most bespoke formatting applied to manuscripts in review is not used for publication as we typeset papers according to Scientific Data'shouse style. For this reason, we do not require or encourage the use of article templates. Authors that require a pre-defined structure are advised to simply copy the headings in section 1 into a blank document and read the rest of this guide for further advice.

Manuscripts published in Scientific Data are not subject to in-depth copy editing. Authors are responsible for procuring copy editing or language editing services for their manuscripts, either before submission, or at the revision stage, should they feel it would benefit their manuscript. Such services include those provided by our affiliates Nature Research Editing Service and American Journal Experts. Please note that the use of such a service is at the author's own expense and in no way implies that the article will be selected for peer review or accepted for publication.

Title & Abstract

Titles must not exceed 110 characters, including whitespaces, and should avoid the use of acronyms (other than very common ones such as DNA, etc), abbreviations, and unnecessary punctuation. Colons and parentheses are not permitted. We recommend only real, technically descriptive words are used to align titles with how users search for information. Do not include dataset brand names in titles, self-constructed acronyms, or words used out of context for the scientific topic, as no user will be searching for them. Capitalise only the first word and any proper nouns and do not include advertising claims not related to the scientific content of the work (e.g. novel, first, AI-ready, open, etc). If unsure, just describe what it is in objective terms.

We recommend the Abstract should not exceed 170 words. It should succinctly describe the data and how it may be used but should not make any claims regarding new scientific findings. We recommend URLs for download, or other details on dataset access, are not included. Please do not use sub-headings to break the Abstract into sections.

Authors & Affiliations

Author affiliations should provide enough detail for the author to be reached, including the department, institution and country where relevant. Full postal addresses are not required. Affiliations should be cited in numerical order within the author list, starting with the affiliations of the first author. Email addresses should be provided for corresponding authors. If you wish to name more than one first author please use the additional footnote such as "These authors contributed equally". All other contributions should be described in the author contributions statement. We do not use other status label footnotes such as "Senior Author".

Background & Summary (Data Descriptors) / Introduction (Articles)

For Data Descriptors, this section should provide an overview of dataset, including the motivation for creating it, as well as outlining the potential reuse value where applicable. Any previous publications that used these data, in whole or in part, should be cited and briefly summarised. Introductions for Articles should provide a similar explanation of why the work was performed and what value you believe it adds. Please note that because Data Descriptors do not present results or analyses there is no formal mandate to cite a given amount of prior art for comparison, however we recommend that at least some other datasets or outputs relevant to the field are cited for readers' general interest. Please do not include subjective claims on novelty, impact, or utility.

Methods

Methods in Data Descriptors should describe the steps or procedures used to create the data, including full descriptions of the experimental design, data acquisition and any computational processing. For secondary datasets compiled from other sources, all details of input data should be described in the Methods (use a sub-heading such as 'Input data' if you wish) at a level of detail that allows readers to source the exact data used (please do not use non-specific URLs or just homepages). Where continuously updated input data have been used, authors should provide the version number or search terms used to create it at a level of detail that allows others to find the same input and repeat the work. Input datasets with DOIs or other formal metadata should be explicitly referenced via our data citation format. For other types without formal metadata, please embed any URLs in the text. Please do not include general results or analyses in this section (or anywhere in the paper) which should solely focus on documenting practical tasks. Where data have been analysed or otherwise published elsewhere then the experimental methods can be cited rather than re-stated, with the new content just focusing on any technical or processing steps additional to what has been previously disclosed.

Articles should describe the full scientific process for how the output or study was generated, but there is less of a requirement to discuss operational aspects like software development or project management processes unless specifically relevant to the science or data science of the piece. Please do not include general descriptions of features or other advertising claims. The focus needs to be on the scientific work (tasks) done.

Papers describing outputs from consortia or other large, multi-stakeholder projects are asked to be mindful of scientific relevance and reader interest9 when describing how they are administered, managed, funded (beyond the practical requirement to state the funder details), or otherwise organised, unless this is of specific relevance to the science.

Data Record (Data Descriptors only)

This section should be used to explain what the dataset contains, including the repository where the dataset is hosted, an overview of the data files and their formats, and any folder structure. Each external dataset should be cited using our data citation format. The section should also define any column headings, variable names, or other fields/metadata if you believe these would not be understood to users (e.g. terms like "gender" or "year" may not need explanation, but any bespoke terms may do). Please do not include summary statistics in this section, but see the description below if you wish to include these.

Data Overview (Optional, Data Descriptors only)

Data Descriptors should not contain results, discussion, or analyses. This is because the fundamental goal of a Data Descriptor is to describe the technical contents of a dataset to encourage others to use it, without bias. Users need details on methods or file descriptions, but if they wish to scrutinise your dataset's contents, we expect them to download and analyse it for themselves. This means we do not recommend researchers include descriptive or summary statistics in their work as any figure, statistic or textural description of this type may be generated from the data file directly.

However, we understand some authors may still wish to do this, so the format supports a short "Data Overview" section for this purpose. Please limit this to a small number of figures or tables (1-2) and a single paragraph of text. Note that the more analysis is shared in the format, the less incentive there may be for others to download, analyse, and cite the dataset themselves. We will request larger amounts are removed or reduced.

Technical Validation (Data Descriptors only)

This section should describe the experiments, analyses or checks needed to support the technical quality of the dataset, with any supporting figures and tables, as needed.

Usage Notes (Data Descriptors only, optional)

'Usage Notes' is an optional section that can be used to provide information that may assist other researchers who reuse your data. Most commonly these are additional technical notes about how to access or process the data. Please do not use this section to write a conclusions section, general selling points, worked cases studies, or similar, as we do not publish these. 

Data Availability

For Articles:

Data Availability statements are required for all Articles. If your Article is sharing a non-code output, such as a dataset, ontology, schema, policy, or any other material that needs to be hosted at external repository, please describe it in the "Data Availability" statement (note it may not always be a “dataset” as such). Please also use this to link out to any platforms, if your paper describes them as its primary output. The statement should include a URL to the source, reference number to a data citation (see guidelines below) and a description of what is practically being shared – best practice is to the name specific files, folder structure, and any information required to understand what they are. In other words, explain what is being shared and where to find it. If there is no output, Data Availability statements are still required, but please state “No data is shared as part of this article” (or similar).

For Data Descriptors:

The key section to describe what data is being shared, and where to find it, is the “Data Record” section. Please ensure this section is detailed enough to allow reader to find and use your dataset.

In addition, we require a short “Data Availability” statement to repeat the main points in the Data Record, meaning some duplication is expected. We suggest repeating the lists of accession numbers, URLs, and the reference numbers for all datasets associated with the paper.

Why is this required, even though Data Descriptors contain descriptions of data in other sections?

We request the additional statement as Data Availability statements (DASs) are a key publication standard. They are present in many other research articles, so we require them to align with wider practice and ensure readers who use them can find information quickly and easily. In addition, some indexing and scraping efforts may look in the DAS discern which datasets are linked to our articles, so the extra heading and information can be additionally helpful.

Code Availability

For all publications, a statement must be included under the subheading "Code Availability" indicating whether and how and custom code can be accessed, including any restrictions to access. This section can also include information on the versions of any software used, if relevant, and any specific variables or parameters used to generate, test, or process the current dataset if these are not included in the Methods. Please see our policy on code availability for more information. The code availability statement should be placed at the end of the manuscript, immediately before the references. 

If no custom code has been used then the statement is still required in order to state this. 

Acknowledgements, Author Contributions, Funding & Competing Interests

The Acknowledgements statement is optional and should contain text acknowledging non-author contributors. Acknowledgements should be brief, and should not include thanks to editors or effusive comments. Please do not include this section if there are no non-author contributors to thank and please do not use it to describe funding. For articles submitted via our new manuscript submission system (SNAPP) this statement is captured by providing information directly on the system. Please do not include the section in the manuscript file as this may be overwritten or excluded if the system information differs. Authors using our old submission system (MTS) need to include the statement in the article file itself, if needed.

The Author contributions statement is required for all articles and should briefly describe each author's contribution to the work. Please see also the Nature journals's authorship policies. For articles submitted via our new manuscript submission system (SNAPP) this statement is captured by providing information directly on the system. Please do not include the section in the manuscript file as this may be overwritten or excluded if the system information differs. Authors using our old submission system (MTS) need to include the statement in the article file itself.

The Competing interests statement is required for all articles. Please see our policies for more information on what may constitute a competing interest. If you are using our new submission (SNAPP) then this statement will be captured via the system. Please do no include an statement in the article file as this may be ignored or overwritten if it differs from information supplied on the online form. Papers declaring "NO" to the competing interests question on SNAPP do not need to provide a statement, as we will add a "No competing interests" statement to the paper for you during typesetting. Authors answering "Yes" will be required to supply a statement explaining the competing interest which will be published instead. For users of old manuscript submission system (MTS) all authors must include a "Competing Interest" sub-heading in the manuscript file with a positive or negative declaration.

The Funding statement is required for all articles. Please note that while funding information is captured on both our manuscript submission systems this is used for internal data capture and compliance and is not published, so all authors (using the SNAPP or MTS submission systems) need to add a 'Funding' sub-heading to the paper listing all relevant funding, organisations, and grant numbers for all relevant projects or authors. If the work is not funded by any external source please still add a statement and declare this.

References

All references should be numbered sequentially throughout the text. Only one publication is given for each number. Only papers that have been published or accepted by a named publication or recognized preprint server should be in the numbered list; preprints of accepted papers in the reference list should be submitted with the manuscript. Grant details and acknowledgments are not permitted as numbered references and no other footnotes should be used in the paper. 

For LaTeX files, please note that references should be embedded directly within the .TEX file. Please do not use separate .bib or .bbl files. If you have created a .tex that requires these, remove the dependency and paste reference list into the .tex directly. Please see the specific notes for LaTeX further down. 

The correct abbreviation for Scientific Data is 'Sci. Data'.

Scientific Data suggests the use of the standard Nature referencing style. See the examples below for a journal article¹, book², book chapter³, preprint⁴, computer code⁵, online material^6-8 and government report⁹. However, please note that the post useful aspect to include are DOIs for all items that have them, as the easiest method for readers to find content. These may be appended to the end of any reference in URL format (https://doi.org/DOI, where DOI is the relevant number). Examples:

1. Schott, D. H., Collins, R. N. & Bretscher, A. Secretory vesicle transport velocity in living cells depends on the myosin V lever arm length. J. Cell Biol. 156, 35‐39 (2002).
2. Hogan, B. Manipulating The Mouse Embryo: A Laboratory Manual 2nd edn (Cold Spring Harbor Laboratory Press, 1994)
3. Haines, N. & Cotter, R. in Studies in Manic Depression Vol. 1 (ed. Boase, N.) Ch. 2 (Oxford Univ. Press, 1982).
4. Babichev, S. A., Ries, J. & Lvovsky, A. I. Quantum scissors: teleportation of single-mode optical states by means of nonlocal single photon. Preprint at https://arxiv.org/abs/quant-ph/0208066 (2002).
5. Gallotti, R. & Barthélemy, M. Source code for: The multilayer temporal network of public transport in Great Britain. Figshare https://doi.org/10.6084/m9.figshare.1249862.v1 (2014).
6. Manaster, J. Sloth squeak. Scientific American Blog Network http://blogs.scientificamerican.com/psi-vid/2014/04/09/sloth-squeak (2014).
7. QGIS Development Team. QGIS Geographic Information System, version 2.18.10. Open Source Geospatial Foundation Project https://qgis.org/en/site/ (2016).
8. Hijmans, R. J., Phillips, S. J., Leathwich, J. & Elith, J. dismo: Species Distribution Modelling https://CRAN.R-project.org/package=dismo (2018).
9. Akutsu, T. Total Heart Replacement Device. Report No. NIH-NHLI-69 2185-4 (National Institutes of Health, 1974).

Citing Data

In line with emerging industry-wide standards for data citation, references to all datasets described or used in the manuscript should be cited in the text with a superscript number and listed in the ‘References’ section in the same manner as a conventional literature reference.

An author list (formatted as above) and title for the dataset should be included in the data citation, and should reflect the author(s) and dataset title recorded at the repository. If author or title is not recorded by the repository, these should not be included in the data citation. The name of the data-hosting repository, URL to the dataset and year the data were made available are required for all data citations. We strongly encourage the use of stable persistent identifiers, such as DOIs, for datasets described in the journal. These should be included in references in a URL format (https://doi.org/XXXXX, where XXXX is the DOI). Please note some repositories may require these be requested in advance. For repositories using accessions (e.g. SRA or GEO) an identifiers.org URL should be used where available. For first submissions, authors may choose to include just the accession number. Scientific Data staff will provide further guidance after peer-review. Please refer to the following examples of data citation for guidance:

1. Zhang, Q-L., Chen, J-Y., Lin, L-B., Wang, F., Guo, J., Deng, X-Y. Characterization of ladybird Henosepilachna vigintioctopunctata transcriptomes across various life stages. figshare https://doi.org/10.6084/m9.figshare.c.4064768.v3 (2018).
2. NCBI Sequence Read Archive https://identifiers.org/ncbi/insdc.sra:SRP121625 (2017). [See note]
3. Barbosa, P., Usie, A. and Ramos, A. M. Quercus suber isolate HL8, whole genome shotgun sequencing project. GenBank https://identifiers.org/ncbi/insdc:PKMF00000000 (2018).
4. DNA Data Bank of Japan https://ddbj.nig.ac.jp/resource/sra-submission/DRA004814 (2016).

[Note]: Please note the SRP accession number should be used, if available, rather than any lower order accession number. This allows the SRA dataset to be cited via a single reference, rather than many.

Figures & Tables

Manuscripts may reference figures (e.g. Figure 1), tables (e.g. Table 1), and Supplementary Information (e.g. Supplementary Table 1, Supplementary File 2, etc.). Please see the additional guidance below for submitting figures, tables and supplementary information, though note they may not be needed for the first round of review. 

Please do not include internal hyperlinks to elements within the manuscript as these are typically removed and replaced with our own linking when typesetting. 

File requirements for first round review

The minimum - and easiest - requirement for first round peer review is to submit a single pdf file for the main article plus any other (pdf) for Supplementary Information, if present. The article file needs to include all figures and tables within it, with captions, and ensure each element is discussed in the main text. If that is done there is no requirement to provide separate figures, tables, or a machine-readable version of the main article for first round review. If we can read everything, we can review it.  

Covering letters

Uploading a covering letter file is a technical requirement for our manuscript submission system that we cannot remove, however there is no requirement to include specific content and covering letters are not used to make editorial decisions on the suitability of works. All submissions that meet our technical criteria are sent for review, regardless of the content in the covering letter, and we do not make decisions based on perceived impact. If there are technical details that you wish to include in a covering letter, please see the guidance below, otherwise any content in the letter is fine as long as you upload a file (even a blank one):

• Suggested reviewers or Editorial Board Members: please include these as you wish, but neither are mandated, and rest assured we have processes to assign manuscripts and reviewers without input, and suggestions are not required to expedite review.
• Non-preferred reviewers or Editorial Board Members: our systems and processes for assigning personnel will check for common conflicts of interest visible from records in the public domain, such as co-authorship of recent papers or shared institutions, however, if you have any live collaborations with researchers that are related to the research topic and may not be discoverable by the editor please let us know.
• Prior publications: These should be disclosed via the relevant question on the submission system, rather than in a letter. If no public link to files is available (e.g. to a published paper or preprint), please make sure we have a copy of the paper within the manuscript files. Please see our policy for further information.
• Notes on affiliations, contact details, or author contributions should all be listed on the system, and additional details may be missed if included in a letter
• Information on the importance of the work or why it should be published will be checked or considered for publication. Reviews are made solely on the content submitted (the article, dataset, and any supplementary files).
• Submission to an article collection: Please ensure this is indicated directly on the system by selecting the relevant collection. If this is not done then it may be missed if included in the letter instead
• Requests for resubmission or appeal: All appeals must be approved via email to scientificdata@nature.com in advance of re-submission, otherwise works duplicated or direclty re-submitted after rejection will be withdrawn.

For revised manuscripts (or Appeals approved by email to re-submit), a separate document is required in all cases, however this should be marked as a 'Response to Reviewers' file on the system (which is visible to all, including reviewers), rather than a covering letter. .

 

4. Submit the manuscript

 

Submit your manuscript and related files via our online system.

5. First round peer review

 

We first perform an initial quality check to check the basic requirements for peer review, such as file availability, data availability, code availability and to check papers are within the scope of the journal. Our team will contact you to fix any issues if required. While we do reject papers outside our technical scope – such as traditional research papers not sharing any data – we send all in-scope manuscripts that meet our technical requirements for peer review and do not make assessments on perceived impact. You will be contacted once initial quality control is complete to inform you peer review has started.

We then send the paper to our Editorial Board, contacting members within the correct subject area to request they handle the paper. Please note their acceptance is dependent on their availability and so it is common to have to contact multiple Editorial Board Members before someone accepts. In the unlikely event we do not have an available Editorial Board Member to assess the paper we will assign it to a member of our in-house staff. We will email you once the handling editor is assigned to inform you who is handling your paper.

The handling Editorial Board Member is then requested to make an initial technical check and commence reviewer invitations. You will be notified by email once this starts, which is the last update we send before any final decision. Please note that sourcing reviewers is typically the longest period of peer review and requires us to invite volunteers to assess the work. The timing of this stage depends on how many reviewers decline or do not respond to our invitations, which determines how many invitations we need to send and how long the process takes. This varies by manuscript and is not something the journal can directly control, however please rest assured we will invite reviewers at the maximum possible rate until we have the minimum two reviewers required to make a quality assessment.

Once the last reviewer is assigned the review invitation process stops and we await the final reports. Reviewers are typically provided 10 days to assess the paper but more time is provided if requested. In rare cases, if commissioned reviews are not submitted after this time, we may go back to the "inviting reviewers" stage to source a replacement.

Once both reviews are received the Editorial Board Member will be asked to assess the reports and make a recommendation - either to Accept, Revise, or Reject the work. We will communicate this decision to all authors via a decision letter alongside the results of our data policy checks, and any remaining requirements.

6. Revised manuscripts

 

What to expect

If your paper receives a "revisions" decision at the first round of peer review you will be invited to revise and re-submit the paper according to the reviewers' and editors' feedback. One month is provided for revisions but we can be flexible if you need more time.

Scientific Data aims for a two-round review process, meaning the first round communicates any potential issues to you and the second-round checks to ensure they have been addressed. If the editor and reviewers are happy with the changes papers may receive an 'Accept' recommendation at the end of round two.

Second round review starts with a quality check of your revised work. This is more detailed than our checks at round one as we aim to ensure the paper is ready for any final publication. Details of our expectations are provided in the next sections.

Once our revised manuscript checks are complete your paper will be sent back to the previous Editorial Board Member who will check your article revisions and response file. In most cases reviewers are invited to re-assess works, however Editorial Board Members are free to make reassessments directly in cases where they feel re-review is not required. The Editor will also assess if their comments from the previous round have been addressed.

Once the process is finished you will receive a second-round decision letter. We will send ‘Accept’ decisions where we feel all major issues have been fixed and the work is technically sound and ready to publish. If not, we may return the paper to you for further revision. Note that while rejection at round two is generally rare it is still used in cases where reviewers feel comments have not been addressed but it would not be advantageous to continue the process.

General requirements

• One copy of the main article file in a machine-readable format (.docx or .tex only - no pdfs - LaTeX users see here). We require a clean copy that is ready for any final acceptance, so please do not include highlighting or other tracked changes, ensure all mandated sections are included, and there are captions for all tables and figures. Tables may be embedded within this if they are machine readable.  
• All figure files, one file per figure. Individual panels a,b,c cannot be supplied separately and need to be merged into one file per figure in the arrangement you want and the letter labels (a, b, c, etc) embedded within the image. 
• Supplementary information document, if relevant (pdf only)
• Any oversize tables larger than 1 A4 page that need to be submitted as Supplementary Tables (in xlsx format or csv)
• A "Response" file with answers to ALL reviewer AND Editor questions (pdf)
• [Optional] a pdf with changes highlighted if you need to show reviewers what was changed and do not feel you can explain this in words. 

Complete guidelines for re-submission and final publication are listed below:

Figures

We recommend Data Descriptors and Articles not have more than eight figures. While this is not a mandate, please note that in many cases the requirement for more figures is due to including general summaries, results, or analyses that may not be within scope of a Data Descriptor. In addition, a limited number of uncaptioned molecular structure graphics and numbered mathematical equations may be included if necessary.

Scientific Data requires authors to present digital images in accord with the policies employed by the Nature-titled journals. 

Authors are responsible for obtaining permission to publish any figures or illustrations that are protected by copyright, including figures published elsewhere and pictures taken by professional photographers. The journal cannot publish images downloaded from the internet without appropriate permission.

Figures should be numbered separately with Arabic numerals in the order of occurrence in the text of the manuscript. Figures presenting quantitative information should include error bars where appropriate and a description of the statistical treatment of error analysis should be included in the figure legend.

Figure lettering should be in a clear, sans-serif typeface (for example, Helvetica) and the same typeface in the same font size should be used for all figures in a paper. Use Symbol font for Greek letters. All display items should be on a white background and in a format that is clear to read. Please do not attempt to accommodate a greater number of data points that you feel a reader can reasonably visually discern. The vertical axis of histograms should not be truncated to exaggerate small differences. Labelling must be of sufficient size and contrast to be readable, even after appropriate reduction. Authors will see a PDF proof that will include figures prior to publication and note that these may be compressed vs the original files. 

Figures divided into parts should be labelled with a lower-case bold a, b, and so on, in the same type-size as used elsewhere in the figure. Lettering in figures should be in lower-case type, with only the first letter of each label capitalized. Units should have a single space between the number and the unit, and follow SI nomenclature (for example, ms rather than msec) or the nomenclature common to a particular field. Thousands should be separated by commas (1,000). Unusual units or abbreviations should be spelled out in full or defined in the legend. Scale bars should be used rather than magnification factors, with the length of the bar defined on the bar itself rather than in the legend. In legends, please use visual cues rather than verbal explanations such as ‘open red triangles’.

Unnecessary figures should be avoided: data presented in small tables or histograms, for instance, can generally be described briefly in the text instead. Figures should not contain more than one panel unless the parts are logically connected; each panel of a multipart figure should be sized so that the whole figure can be reduced by the same amount and reproduced at the smallest size at which essential details are visible.

Figures for peer-review

At the initial submission stage authors may choose to upload separate figure files or to incorporate figures into the main article file, ensuring that any inserted figures are of sufficient quality to be clearly legible. When submitting a revised manuscript all figures must be uploaded as separate figure files ensuring that the image quality and formatting conforms to the specifications below.

Figures for publication

When creating and submitting final figure files, please follow the guidelines below. Failure to do so can significantly delay publication of your work.

Each complete figure must be supplied as a separate file upload. Multi-part/panel figures must be prepared and arranged as a single image file (including all sub-parts; a, b, c, etc.). Please do not upload each panel individually. This is because our typesetters will not arrange them for you, so if separate panels are used please combine them in a suitable image editor and re-save the full figure as a single image prior to submission. 

Authors are asked to provide figures of a sufficient resolution for final online publication, however, please do not upload figures that are excessively large. As long as the image is legible it will be suitable for peer review and publication, noting that our typesetting process will compress files to standard quality for web and pdf publication anyway. Figure file sizes larger than a few Mbs are likely codifying information or resolution that will not be retained. 

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figure legends begin with a brief title sentence summarizing the purpose of the figure as a whole and continue with a short description of what is shown in each panel and an explanation of any symbols used. Legends must total no more than 350 words and may contain literature references.

Any descriptions too long for the figure legend should be included in the Methods section. 

Tables

Authors may provide tables within the Word document or as separate files. Legends, where needed, should be included in the Word document. We recommend Data Descriptors and Articles should no more than ten tables, but more may be allowed if needed. Tables may be of any size, but only tables that fit onto a single printed page will be included in the PDF version of the article. Please do not use complex nested or merged cell structures and we recommend a standard 2D row/column format. If the information depicted is not numerical please consider whether a table is needed vs some other format (e.g. text, a list, etc).  

Oversize tables

Due to typesetting constraints, tables that cannot be fit onto a single A4 page cannot be included in the PDF version of the article and will be hosted as Supplementary Tables, although please note that table sizes are generally smaller in publications than in draft (meaning up to 1.5 pages in draft may possibly fit). Any oversize tables must be labelled in the text as ‘Supplementary' tables and numbered separately from the main table list e.g. ‘Table 1, Table 2, Supplementary Table 1’ etc. Please note bibliographic references cannot be included within Supplementary Tables and should not be listed in the reference list, which only refers to references used in the main article file. They also cannot have captions in the main article as they are not included within it, but are hosted separately. If you do wish to include any context or information in the supplementary table please find another means of mentioning these references on the main text.

Finally, please note that especially large tables (e.g. 100s or 1000s or rows) should not be shared with the manuscript at all but would be better considered as data tables and deposited within your repository-deposited dataset instead. Please do not call these "supplementary" as this causes confusion with materials hosted alongside the paper. Please choose a descriptive file main and refer to the item in the text as being available at the repository (with a URL and/or data citation to allow readers to find it).

 

Equations

Equations and mathematical expressions should be provided in the main text of the paper. Equations that are referred to in the text are identified by parenthetical numbers, such as (1), and are referred to in the manuscript as ‘equation (1)’.

Supplementary information

Scientific Data discourages authors from supplying text, figures or tables as supplementary files. As much as possible, these types of content should be included in the main manuscript. 

The main sections of the Data Descriptor manuscript, and particularly the Methods section, have no word length limits, and the journal is not printed, so unless the supplementary information document would extend the length of the paper significantly (e.g. by 10 pages or more) we recommend the content is included in a single article file and not split out. Please do not use Supplementary Information to include information that is out of scope for the article type chosen, such as to add results, discussion, or analyses to your Data Descriptor. The general expectation is that these are additional methods, validation, data dictionaries, or technical notes (such as how to run the code) that simply won't fit. 

The guidelines below explain how to generate Supplementary Information (SI) where it cannpot be avoided. While we accept two types of SI - documents and oversize tables - instructions on the second one are explained in the previous section. 

1.  A supplementary information document is a pdf containing any supplementary text, figures, small tables or references if these elements are too numerous to fit in the main text. These should always be self-contained and submitted as a single pdf, so please do not supply individual items. Supplementary information documents are not typeset and are published unchanged from the format and content you supply, so we do not need to collect their individual components, unlike for the main paper. Items in the file should be numbered in the document and this numbering should be separate from that used for regular tables and figures appearing in the main article. Please use the formalism "Figure S1", "Table S1", etc to distinguish these elements. 
2. Refer to the supplementary information file in the main paper, either referring to the whole document (the supplementary file), or individual items (e.g. supplementary figure 1). Please do not use the word 'supplementary' to refer to any externally hosted file (e.g. a dataset hosted within a repository) or items included in the main paper as this causes confusion during typesetting and publication for where to find the files. This is important to note if the Editor requests files are removed from SI and uploaded to a repository - please rename them if so. 
3. Remember to include a brief title and legend for items within the supplementary information document.
4. File sizes should be as small as possible, with a maximum size of 10 MB, so that they can be downloaded quickly. We mandate pdf formats for supplementary documents and xlsx or csv files for oversize supplementary tables (see above)
5. When supplying multiple supplementary figures, they should be merged into the main supplementary information file, with figure legends immediately below each figure. A table of contents should be included on the first page, listing the page number of each supplementary figure. Do not share separate supplementary figure files as these are not typeset, as outlined above
6. References in Supplementary Information documents should start again from 1 in a new numbered listed from the main article. Please note these will not be formally indexed by bibliographic indexing services and will likely not count (be trackable) as formal citations effecting user-level metrics such as h-indices. Please consider this - that no formal credit is possible - when using supplementary references and please always consider if the reference could not be included in the main text instead. 

 

Instructions for LaTeX users

Note these are instructions are only required revised manuscripts, where we require an editable version in preparation for typesetting. Articles submitted for first round peer review may be supplied in pdf format (read only) for simplicity. 

For revised manuscripts, the core requirement is to supply a single .TEX file marked as an "Article" on the system. This should be standalone, meaning it can be compiled to a pdf without any additional .bib files, style sheets, or other dependencies.

To check this, please open the .TEX within your editor without providing read access to any other files. If it fails to compile, or sections are missing, please check to ensure it is truly standalone and no dependencies are required. If that has been checked, uploading the .TEX file to our submission system (as an "Article") without other files should allow a merged pdf to be generated. Note the most common issue is the reference list is missing due to a reliance on a separate .bib/bbl. Please copy the reference list from the .bbl file, paste it into the main manuscript .tex file, and delete the associated \bibliography{sample} command. 

Please do not compile your own pdf and upload it to system as we then have no guarantee the files will match, or confirmation the .tex will compile. Similarly, .zip files of associated files should also not be uploaded as the .tex file should be possible to compile to a complete, viable pdf without the requirement for additional files.

We do not provide, suggest or recommend the use of a LaTeX template so if you find a previous or legacy version of this via platforms such as Overleaf please do not use them as they may be out of date. Please do not use the LaTeX template of any other journal.

The only formatting requirement are the headings listed in section 1, so copying these into a blank document and using any legible formatting stype is acceptable for first and revised submissions. We recommend Computer Modern fonts. Non-standard fonts should be avoided. For the inclusion of graphics, we recommend graphicx.sty. Please use numerical references only for citations, and include the references within the manuscript file itself as explained. Exact reference formats are not important as long as these contain all the key information (a name, a title, a journal, and - most importantly - a DOI).

Please do not include any internal hyperlinks to sections, headers, figures or tables, via \cref, or similar. Just refer to these elements using free text (e.g. 'see Fig. 1', without linking). This is because all internal links are set by our house style and are created during typesetting, rather than using markup supplied in the .tex file.

Consortia authorships

​A consortium author is the name of a group, rather than an individual, in the author list. These should only be used if a standard author list, naming all contributors directly in the paper, would be considered excessively long (e.g 100 or more people). Consortium authorships should not be used for small groups that could easily be accommodated into the main list, or for advertising or promoting projects. 

If a consortium is included in the main author list, all members of the consortium are considered bona fide authors, and must be listed together with their affiliations at the end of the Author Contributions statement. The authors and affiliations for the consortium members are an extension of the main author list, used simply because there is no space for them at the start of the paper. Therefore any affiliations already included in the main author list should not be repeated in the Author Contributions statement and the numbering of the affiliations in the consortium should continue in numerical order from those in the main author list – they should not start again from 1. If a member of the consortium already appears as an individual name in the main author list, then his/her affiliations should be identical in the consortium author list. The consortia itself should be acknowledged with the footnote "A full list of members appears in the Author Contributions". If you need to give credit to a consortium, a project or a group of people who do not meet authorship criteria, you can add a mention in the Acknowledgements section or elsewhere (in which case, a full list of members can be provided as a Supplementary Note in the Supplementary Information, if desired). Please do not use consortia authorship to credit individuals who have not actively participated in the creation of the paper or new dataset. If you are submitting a description of a secondary dataset compiled from previously published input data then authorship should not be typically granted to the data owner unless they have been actively involved in the (new) processing or compilation to create the secondary output. Instead, individuals should be credited via formal citation of the existing dataset, paper, or any other relevant works, in the same way as using any piece of prior art.