- Image and Text Integrity
- Conflict of Interest
- Prior Publication and Presentations at Meetings
- Unpublished Data and Personal Communications
- Related Papers
- Security and Safety Concerns
- TOP Guidelines
- Statistical Analysis
- Study Design
- Guidelines for Specific Types of Studies
- Data and Code Deposition
- Data and Materials Availability after Publication
- Copyright and License to Publish
- Access Policies
- Embargo Policy
- Corrections, Expressions of Concern and Retractions
This section outlines our Editorial Policies for all Science Journals. For journal-specific Information for Authors follow the appropriate link:
Authors in Science journals must fulfill the criteria described below, which are informed by the International Committee of Medical Journal Editors (ICMJE) definition of authorship. Specifically,
- Each author is expected to have made substantial contributions to the conception or design of the work;
- OR the acquisition, analysis, or interpretation of data;
- OR creation of new software used in the work;
- OR have drafted the work or substantially revised it;
- AND has approved the submitted version (and any substantially modified version that involves the author’s contribution to the study);
- AND agrees to be personally accountable for the author’s own contributions and for ensuring that questions related to the accuracy or integrity of any part of the work, even ones in which the author was not personally involved, are appropriately investigated, resolved, and documented in the literature.
Exclusion from authorship of individuals who have made author-level contributions is not permitted for papers published in Science journals, nor is guest or honorary authorship. In addition, artificial intelligence tools cannot be authors. Other individuals who have participated in generation of the research paper but who do not meet the criteria for authorship should be listed in the acknowledgments section with a brief indication of the nature of their contribution.
In addition, corresponding authors must:
- Ensure that all listed authors have received and approved the manuscript prior to submission.
- Receive all substantive correspondence with editors, as well as full reviews.
- Verify that all data, materials (including reagents), and code, even those developed/provided by other authors, comply with the transparency and reproducibility standards of both the field and the journal.
- Ensure that original data/materials/code upon which the submission is based are preserved and retrievable for reanalysis.
- Confirm that the presentation in the paper of the data/materials/code accurately reflects the original sources.
- Foresee and minimize obstacles to the sharing of data/materials/code.
- Ensure the entire author group is fully aware of and in compliance with best practices.
When a new manuscript is submitted to a Science journal, all listed authors will receive an email with a link to confirm authorship. Submission of a paper that has not been approved by all authors may result in immediate rejection without appeal. Any addition or removal of authors from a paper or change in their listed order must be approved in writing by all of the original authors.
For manuscripts that are accepted, all authors are required to explain their contribution to the manuscript; agree to the conditions of publication, including the availability of data, code, and materials; and declare any conflicts of interest. The senior author from each group is required to have examined the raw data that their group has produced. We encourage all authors to state their contribution to the study in the acknowledgments section following the CRediT model; this information will be published in the paper.
The scientific record is indexed for posterity by author names, but authors may have occasion to change their names for a variety of reasons (e.g., gender identity or change in marital status). As such, we have updated our policies to modify an author’s name accordingly in that author’s prior publications across the Science journals. Authors should email [email protected] with citations to the papers they would like us to modify (including titles and DOIs for each paper) and an ORCID iD associated with the new name. There is no need to justify the request or to alert co-authors. Our default policy will then be to change any or all the author’s given/middle/family names and initials in the full text (HTML), PDF, and metadata records of those papers, with no other indication or notification that the changes have been made. If an author prefers to publicize a name change, we can discuss the options available. Only the new name will be associated with the author’s papers in the search functionality on the Science journals’ websites. We will also work with Crossref to help ensure that future citations of the author’s papers use the new name across the Science journals and in other journals whose publishers rely on Crossref to proofread their reference lists.
Plagiarism. Plagiarism is considered by the US Office of Research Integrity to “include both the theft or misappropriation of intellectual property and the substantial unattributed textual copying of another's work.” Substantive re-use of text or figures previously published by a current author without appropriate citation is considered self-plagiarism. The Science journals use the Crossref Similarity Check service to detect plagiarism and follow guidelines from the Committee on Publication Ethics (COPE) in dealing with cases of suspected plagiarism.
Artificial intelligence (AI). Text generated from AI, machine learning, or similar algorithmic tools cannot be used in papers published in Science journals, nor can the accompanying figures, images, or graphics be the products of such tools, without explicit permission from the editors. In addition, an AI program cannot be an author of a Science journal paper. A violation of this policy constitutes scientific misconduct.
Images. Images presented in research papers should correctly represent the original data. No part of a digital image may be selectively manipulated or altered. When figures are assembled from multiple images or non-contiguous portions of the same image, a line or space should indicate the border between the separate parts. Science journal editors may use image proofing software to screen images.
Authors. Every author of content submitted to a Science journal is required to disclose affiliations, funding sources, and competing interests that might be perceived as sources of bias related to the reported research and/or presented content. Specific details about the required disclosures are listed below, along with information about placement of these disclosures within the article. This policy applies to research, reviews, and opinion-type articles.
Science Journals require disclosure of:
1. Institutional affiliations
The title page of every submission must include, for all authors, the academic, corporate, government, industry, and/or other relevant institutional affiliations where the work was performed.
2. Funding and contributions
All authors must disclose complete and correct information about any and all financial contributions to the work being reported. This information should be listed in the Funding statement in the Acknowledgments section of the manuscript to ensure transparency during the review process and will be included in the final published work. Non-monetary (in-kind) contribution of goods or services may, if appropriate, be cited in published acknowledgments in the paper.
3. Competing interests
All authors must disclose complete and correct details of competing interests that have occurred within 5 years of inception of the research or clinical study under consideration. Interests outside of the 5-year time frame must also be declared if they could reasonably be perceived as competing. When in doubt, authors should disclose the relationship. This information should be summarized in a Competing interests statement in the acknowledgments section of the final published work. Authors can provide a URL to a list of an author’s affiliations/interests/relationships in addition to the Competing interests statement.
A competing interest exists when one’s professional judgment about the execution of the research and/or the presentation of the content is, or could reasonably be perceived to be, influenced by other interests. Review the examples below in the context of the 5-year time frame mentioned above when preparing your Competing interests statement.
Financial competing interests include but are not limited to:
- Awarded, planned, or pending patents, including individual applications or those belonging to the institution to which the authors are affiliated and from which the authors may benefit
- Ownership of stocks, shares, or stock options even if not publicly traded
- Paid employment or consultancy
- Received payment for serving as a member of an Advisory Committee and/or as an officer/member of the Board for any entity engaged in activity related to the subject matter of this contribution
Non-financial competing interests that could impact the research reported here include but are not limited to:
- Acting as an expert witness
- Member of a government or other advisory board
- Relationship (paid or unpaid) with organizations and funding bodies, including nongovernmental organizations, research institutions, or charities
- Membership in lobbying or advocacy organizations
- Writing or consulting for a company whose activities might impact the objectivity of this paper
- Personal relationships (e.g., friend, spouse/partner, family member, mentor, adversary) that could affect objectivity
- Personal beliefs (political, religious, ideological, or other) related to a paper’s topic that might interfere with an unbiased publication process (at the stage of authorship, peer review, editorial decision-making, or publication)
Reviewers. Reviewers form the cornerstone of the peer-review process, and their evaluations ensure the quality of published research. Therefore, the editors seek reviewers for Science journals who do not have conflicts of interest with the authors or reported research in the manuscripts they read. In addition to this precaution, reviewers are required to disclose any conflicts with the evaluation of the paper, and this information is taken into account by the editors when decisions are made. See here for additional ethical guidelines for Reviewers.
Professional Editors. Editors for Science, Science Immunology, Science Robotics, Science Signaling, and Science Translational Medicine are required to fill out a conflict of interest form, which is then evaluated by Science journals/AAAS management. Professional editors for the Science journals may not have any financial or management interest in any biotechnology, pharmaceutical or biomedical or engineering device, institution, or company.
Academic Editors. Editors for Science Advances and Science Robotics are required to disclose any conflicts with the evaluation of the paper and may be required to recuse themselves based on conflicts of interest.
Advisors & Reviewing Editors. Members of Science journals’ advisory boards may be consulted by the staff editors at the initial evaluation of newly submitted manuscripts. They are required to declare any conflicts pertaining to particular submissions.
The Science journals will not consider any original research paper or component of a research paper that has been published or is under consideration for publication elsewhere. We do not regard dissertations/theses as prior publications. We do regard peer-reviewed and citable conference proceedings as prior publication. We support posting of research papers on preprint servers such as arXiv.org, bioRxiv.org, ChemRxiv.org, medRxiv.org, and the like. Distribution on the internet in other ways may be considered prior publication and may compromise the originality of the paper as a submission to a Science journal. Please contact the editors with questions regarding allowable postings to other servers. While a manuscript is under consideration at a Science journal, no versions revised in response to editorial input and peer review should be posted on a preprint server. Our preprint policy is intended to facilitate communication and exchange of information between scientists. We strongly believe that coverage in the popular press is more appropriate at the time of publication once peer review is complete and the final version is available.
We likewise generally encourage presentation of original results prior to publication at scientific meetings but discourage authors from overtly seeking media coverage ahead of peer review and final publication. If a reporter attends an author’s session at a meeting and writes a story based only on the presentation, such coverage will not affect a Science journal’s consideration of the author's paper.
Citations to unpublished data and personal communications cannot be used to support significant claims in the paper. Papers will be held for publication until all “in press” citations are either published or posted in full on a preprint server.
Copies of any papers by any subset of the authors that are related to the submitted manuscript and are under consideration or in press at other journals must be included with the submission of the original and/or revised version of the paper. Related papers posted at a preprint server must be cited. While your paper is under consideration at a Science journal, please contact your editor if a related paper is submitted elsewhere. Access to this material will let the editors put the results in the submission in the proper context and make the best decision regarding the submitted manuscript. Failure to disclose related papers may lead to rejection.
Authors and reviewers are expected to notify editors if a manuscript could be considered to report Dual Use Research of Concern (DURC). Gain-of function experiments are considered under our DURC policies. The National Science Advisory Board for Biosecurity has defined DURC as “life sciences research that, based on current understanding, can be reasonably anticipated to provide knowledge, information, products, or technologies that could be directly misapplied to pose a significant threat with broad potential consequences to public health and safety, agricultural crops and other plants, animals, the environment, materiel, or national security.” Papers identified as possible DURC risks, or analogous risks in the physical sciences, will be brought to the attention of the Editor-in-Chief for further evaluation. If necessary, outside reviewers with expertise in the area will be consulted.
Safety considerations are essential in the planning of scientific experiments. Authors should clearly indicate in the Methods section any acute hazards associated with particular chemical reagents or products (such as extreme toxicity or explosion risks), as well as the appropriate mitigation steps taken. The American Chemical Society offers helpful guidelines here. Guidelines for working with hazardous biological materials are provided in the Centers for Disease Control and Prevention / National Institutes of Health publication, Biosafety in Microbiological and Biomedical Laboratories, 6th Edition.
The Science journals support the Transparency and Openness Promotion (TOP) guidelines to raise the quality of research published in Science and to increase transparency regarding the evidence on which conclusions are based.
1. Citation standards
All data, program code, and other methods must be appropriately cited using digital object identifiers (DOIs), journal citations, or other persistent identifiers. Exceptional circumstances requiring special treatment, such as proprietary information requiring nondisclosure agreements, should be discussed with the editor no later than at the manuscript revision stage and spelled out explicitly in the acknowledgments.
2. Data standards
All data used in the analysis must be available to any researcher for purposes of reproducing or extending the analysis. Data must be available in the paper or deposited in a community special-purpose repository or a general-purpose repository such as Dryad (see Data and Code Deposition). Exceptional circumstances requiring special treatment, such as protection of personal privacy or purchase of datasets from third-party sources, should be discussed with the editor as early as possible (no later than at the manuscript revision stage) and spelled out explicitly in the acknowledgments. Problems in obtaining access to published data are taken seriously by the Science journals and can be reported at [email protected].
3. Analytic methods (code) transparency
In general, all computer code central to the findings being reported should be available to readers to ensure reproducibility. If the software used is commercially available or the source code is already publicly archived, it should be referenced in an appropriately formatted citation (with the version included, if necessary for accurate replication). Author-written source code that is not yet publicly available should be archived in a permanent public repository prior to publication and likewise cited (see Data and Code Deposition). In exceptional cases where for example, security concerns, legal restrictions, or proprietary hardware preclude sharing of custom code, an alternate means of ensuring reproducibility must be arranged with the editor. Our preferred option in such exceptional cases is for the Materials and methods section to include pseudocode that fully clarifies the underlying algorithms; this pseudocode will be subject to peer review and may require further elaboration in accord with reviewers’ feedback. As with data, the reason(s) for the code restrictions in these special cases should be explained clearly in the acknowledgments.
4. Research materials transparency
The Science journals require the materials and methods section to provide sufficient detail to allow replication of the study. Study design should be described in detail, and reagent description should facilitate replication (for example, source and purity should be specified, there should be evidence that antibodies have been validated, cell lines should be authenticated, and RRIDs should be provided when available). Statistics must also be comprehensively described, as outlined below.
Materials/samples used in the analysis must be made available to any researcher for purposes of directly replicating the procedure. Authors are also expected to honor reasonable (consistent with community standards) requests for research materials/samples to the extent feasible, so that other research groups can extend and advance the results. Any restrictions on the availability of materials, and likewise any pertinent Material Transfer Agreements (MTAs), must be disclosed to the editor no later than at the manuscript revision stage and spelled out explicitly in the acknowledgments. Fossils or other rare specimens must be deposited in a public museum or repository and available for further research.
5. Design and analysis transparency
The Science journals encourage authors to follow relevant standards for their field for reporting key aspects of the research design and analysis. Authors should report which standards were followed and note any deviations from the guidelines. Where applicable, note which guidelines were followed.
A chart providing links to major biomedical research reporting guidelines is available at the U.S. National Library of Medicine. Many of these can be found at the EQUATOR website. Other guidelines may be found at the relevant society website: for example the American Chemical Society gives standards for the characterization of organic and inorganic molecular compounds.
See Guidelines for specific types of studies below.
6. Preregistration of studies
If the study that is submitted to a Science journal for publication was preregistered, such as at the Center for Open Science’s Open Science Framework or ClinicalTrials.gov, authors should provide a link to the registration of the study upon submission of the paper to a Science journal.
7. Preregistration of analysis plans
If the analysis plan was preregistered in an independent institutional registry, authors should provide a link to the registration of the analysis plan upon submission of the paper to a Science journal.
The Science journals encourage the submission of replication studies that provide new insights into previously published results. We hold replication studies to the same standards as other content submitted to the journals.
Authors should describe statistical methods with enough detail to enable a knowledgeable reader with access to the original data to verify the results.
- Data pre-processing steps such as transformations, re-coding, re-scaling, normalization, truncation, and handling of below-detectable-level readings and outliers should be fully described; any removal or modification of data values must be fully acknowledged and justified.
- Descriptive statistics should be presented for variables that are integral to subsequent analyses and interpretation of the study findings.
- The number of sampled units, n, upon which each reported statistic is based must be stated.
- For continuous variables, distributions should be described using graphical displays such as scatterplots, boxplots, or histograms, or by reporting measures of central tendency (e.g., mean or median) and dispersion (e.g., SD, interquartile range).
- For continuous variables that are approximately normally distributed, mean and SD are suitable measures for center and dispersion, respectively.
- For continuous variables with asymmetrical distributions, median and range (or interquartile range) are preferred to mean and SD.
- All measures of central tendency or dispersion that are used should be identified.
- For very small samples sizes (e.g., n< 20), presentation of all data values in tabular format is desirable unless presentation would violate restrictions for privacy or confidentiality for human subjects.
- Units should be supplied for all measurements.
- Methods used for conducting statistical tests (e.g., t test, Wilcoxon signed rank test, Wald test of regression coefficient) and for constructing confidence intervals (e.g., normal-based 95% CI: mean ± 2 SD, likelihood ratio–based interval) should be clearly stated. Mention methods used in the materials and methods section and then provide the individual test name in the figure legend for each experiment.
- The testing level (alpha) and whether one-sided or two-sided testing was used should be reported for each statistical test; typically, two-sided testing is appropriate, but if one-sided testing is used, its use should be justified.
- Adjustments made to alpha levels (e.g., Bonferroni correction) or other procedure used to account for multiple testing (e.g., false discovery rate control) should be reported.
- When Bayesian analyses are conducted, any assumptions made for prior distributions must be fully described.
- Sufficient information should be supplied to allow readers to judge whether any assumptions necessary for the validity of statistical approaches (e.g., data are normally distributed, survival data are consistent with proportional hazards in a Cox regression model) have been verified.
- An accounting of missing data values should be provided; if imputed data values are used in statistical analyses, the methods used for imputation should be fully described.
Authors should present results in a complete and transparent fashion so that stated conclusions are backed by appropriate statistical evaluation and limitations of the study are frankly discussed.
- Point estimates of population parameters (e.g., mean, correlation coefficient, slope) or comparative measures (e.g., mean difference, odds ratio, hazard ratio) should be accompanied by a measure of uncertainty, such as a standard error or a confidence interval.
- Results of each statistical test should be reported in full with the value of the test statistic and P value, and not simply reported as significant or non-significant; more than two significant digits on P values are usually not needed except in situations of extreme multiple testing (e.g., in genetic association studies where stringent corrections for multiple testing might be used).
- Any results that are reported to constitute a blinded, independent validation of a statistical model (or mathematical classifier or predictor) must be accompanied by a detailed explanation that includes: 1) specification of the exact “locked down” form of the model, including all data processing steps, algorithm for calculating the model output, and any cutpoints that might be applied to the model output for final classification; 2) date on which the model or predictor was fully locked down in exactly the form described; 3) name of the individual(s) who maintained the blinded data and oversaw the evaluation (e.g., honest broker); and 4) statement of assurance that no modifications, additions, or exclusion were made to the validation dataset from the point at which the model was locked down and that neither the validation data nor any subset of it had ever been used to assess or refine the model being tested.
In clinical and preclinical studies, the following components should be specified before the initiation of the experiments or trial. In the first section of the Materials and Methods, which should be titled “Study Design,” state how you have addressed each of these points (if applicable):
- Sample size. How did you select your sample size? Did you use a power analysis to calculate the sample size necessary to achieve a reliable measurement of the effect? Did you alter this number during the course of the study and, if so, why?
- Rules for stopping data collection. Did you define rules for stopping data collection in advance (for example, specific intermediary and final endpoints)?
- Data inclusion/exclusion criteria. What criteria did you apply for inclusion and exclusion of data? Were these criteria established prospectively?
- Outliers. How were outliers defined and handled? Were they defined before the beginning of the study? Have you reported outliers that were excluded?
- Selection of endpoints. Were the primary and secondary endpoints prospectively selected? If multiple endpoints were assessed, were the appropriate statistical corrections applied.
- Replicates. How many times was each experiment performed? How were the number and composition of replicates determined? Specify both sampling and experimental replicates. Were the results substantiated by repetition under a range of conditions?
In addition, the Study Design section must describe how and why the study was conducted and how the data were collected. Specifically,
- Research objectives. State the objectives of the research, clearly distinguishing pre-specified hypotheses from hypotheses suggested after initiation of the data analyses.
- Research subjects or units of investigation. Describe the type of research subjects (e.g., cancer patients, healthy volunteers), animals, or experimental units (e.g., cell cultures) studied.
- Experimental design. Describe the overall design (e.g., randomized controlled clinical trial, controlled laboratory experiment, observational study, survey). Include the treatments that were applied, the types of observations made and the measurement techniques used. The details of the measurement methods should be described in separate sections. If a questionnaire was used to obtain information from human subjects, include it as part of the supplementary materials.
- Randomization. Include in the description of the study whether the subject or other experimental units were assigned randomly to the various experimental groups and, if not, how the sample was selected (e.g., random sample, stratified sample, matched case-control sets). The population from which the sample was taken should be specified. How was randomization performed? Were the data collected and processed randomly or were they grouped?
- Blinding. Include whether the study was blinded and the method used for allocation concealment, blinded conduct of the experiment, and blinded assessment of outcomes. Was the investigator aware to which group a particular animal taken from a cage (for example) was allocated? Were the animal caretakers and investigators conducting the experiments blinded to the allocation sequence? Were the investigators who assessed, measured, or quantified the results blinded to the intervention?
Studies that involve animals
All use of animals in research (both laboratory and field based) should be ethical and humane and we support the requirements of the United States Animal Welfare Act. We encourage authors to explore and practice methods that follow the 3Rs, first introduced in 1959, as required by law in the United Kingdom (Directive 2010/63/EU) and summarized as:
1. Replacement – use of substitutes for animals in a study when they exist;
2. Reduction – use of methods that enable researchers to obtain comparable levels of information from fewer animals, or to obtain more information from the same number of animals;
3. Refinement – use of methods that alleviate or minimize potential pain, suffering, or distress, and enhance animal welfare for the animals used.
We encourage authors to ensure that they have obtained the most up-to-date recommendations for humane animal use and care. Authors should describe how their work addresses these requirements in the supplementary materials. In addition, authors should meet the requirements below specific to the context of the study.
Laboratory animals. For all laboratory animal experimentation described in the manuscript, we require that authors state in the materials and methods their adherence to the NIH Guide for the Care and Use of Laboratory Animals or the equivalent, and that they obtained the appropriate Animal Care and Use approvals from their institution. Species, strain, sex, age, origin, care (including pain reduction and humane euthanasia procedures), and housing of laboratory animals should be detailed in the main text or supplementary materials.
Wild, or temporarily captive animals. Researchers should speak to the ethics, humane treatment, and the 3Rs of animals used during the study. Methods and approaches should meet the requirements of the United States Animal Welfare Act or the equivalent. Researchers must state that the work has been approved by their institutional Animal Care and Use committee and should strive to meet all requirements recommended by scientific societies focused on the specifics of their study organism(s) (e.g., mammalogical, ornithological, herpetological, ichthyological, etc.). In addition, collection of samples and specimens must meet further requirements, as laid out below in Studies involving field collection of samples, specimens, or fossils.
Genetically modified animals. Authors must meet the above-stated care and use requirements for laboratory animals. In addition, to avoid confounding effects of inbred strain background, littermate controls should generally be used, although exceptions may be allowed. Justification for other control animals should be included. Authors should fully describe the source of their animals and number of times backcrosses were performed.
Studies involving the field collection of samples, specimens, or fossils
The Science Journals follow the intentions laid out in the Nagoya Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from their Utilization to the Convention on Biological Diversity, the intention of which is to ensure that the sharing of benefits arising from the utilization of genetic (biological) resources is executed in a fair and equitable way and requires authors to follow these guidelines. Along these lines, we require that legal permits from the country of origin must have been obtained, and be available for review, for all research involving the collection of biological specimens, samples, or fossils. Specimens and fossils should be deposited in a museum, or otherwise fully available for researchers to access, and accession information must be provided. We do not allow for specimens or fossils to be deposited in private collections unless full public access is available and accession information provided.
Paleontological and archaeological studies
All fossils, artifacts, ecofacts, and human remains (including ancient DNA) should be collected in an ethical and equitable way. Along these lines, legal permits, from the country of origin, must have been obtained and be available for review. We encourage the involvement of local preparators and experts and encourage such contributors to be recognized as authors when their contributions are significant (e.g., fossil acquisition and preparation). Special attention should be paid to fossils collected from regions where inequity and abuse has been identified (e.g., Burmese amber from 2017 onward), and evidence demonstrating that these practices have not been involved in production of the fossil(s) should be described in the paper and available on request. We also require that best practices around verification and authentication be followed, as described in the Best Practice Guidelines for Repositing and Disseminating Contextual Data Associated with Vertebrate Fossils from the Society of Vertebrate Paleontology.
To ensure that these guidelines are followed, and that editors, reviewers, and readers can easily evaluate authenticity and ethical acquisition, the Science journals require the inclusion of a specific “Provenance” section within the supplementary materials that clearly states: 1) where, when, and by whom the fossils, artifacts, ecofacts, and/or human remains were collected; 2) how, and by whom, the fossils, artifacts, ecofacts, and/or human remains were validated for authenticity and aged/dated; and 3) where the fossils, artifacts, ecofacts, and human remains will be housed and/or how they can be freely accessed.
Human subjects research
Informed consent must be obtained for studies on humans after the nature and possible consequences of the studies are explained. A statement that informed consent was obtained must also appear in the manuscript. All research on humans must have approval from the IRB (Institutional Review Board) or an equivalent body. The editors reserve the right to request IRB documents associated with a particular paper. Gender and age of all subjects should be provided in the main text or supplementary materials.
A CONSORT statement includes recommendations, a checklist of items that should be included in a comprehensive report, and a participant flow diagram. The recommended checklist should be completed and provided at the time of manuscript submission. The recommended trial flow diagram may be presented as a figure (usually Fig. 1). Reports of randomized controlled trials that do not conform to the CONSORT guidelines may be returned to authors for revision prior to formal review.
Registration of clinical trials. Clinical trials should generally be registered in accordance with the criteria outlined by the ICMJE, including the June 2007 update. Authors should provide the trial registration number in the acknowledgments section and provide a link to the trial registration, to be cited as a reference.
Putative biomarkers must be evaluated with an independent validation set/cohort. Reports of unvalidated biomarkers will only be considered in the context of a clear experimental, mechanistic connection to disease or other unique contribution to understanding of disease or clinical practice. A statement should be included in all biomarker papers describing how overfitting (training models on large numbers of variables measured on small numbers of subjects) and other forms of bias were avoided. We strongly recommend that all papers reporting potential new biomarkers be evaluated by an independent statistician before submission.
Computational models should be validated either experimentally or through a dataset independent of the training set. All assumptions should be clearly stated, with sources provided in the references and notes section. (See Research standards for code availability requirements)
Small molecule studies
All studies that make use of chemical compounds (including pharmaceutical research) must disclose the full structures of those compounds, including stereochemistry if known. If the compounds are not readily available from commercial sources, the methods used for their preparation, purification, and characterization must be disclosed in full detail. 1H and 13C nuclear magnetic resonance spectra should be included in supplementary materials for all synthesized organic compounds.
As outlined in the TOP guidelines above, the Science journals generally require all data underlying the results in published papers to be publicly and immediately available. Post-publication embargoes are not permitted, nor are stipulations for readers to contact the authors (rare exceptions involving third-party datasets must be discussed with the editor prior to publication and explained in detail in the acknowledgments). Community standards for what constitutes raw data continue to evolve, but at the very least, when datapoint values are not trivially discernible in plots, the tabulated underlying data should be archived in a standard machine-readable format (e.g., csv, tsv, json, or xml) and should be accompanied by a brief explanation of their structure and meaning. Compliance with MIBBI guidelines (Minimum Information for Biological and Biomedical Investigations) is encouraged. The Science journals support community-driven efforts to collect and cross-reference data systematically in field-specific repositories, and specific guidelines are elaborated below. In cases where datasets do not yet fall under the purview of a dedicated subject repository, they should be archived in a permanent, independent repository such as Dryad, Dataverse, or Zenodo. Institutional repositories that provide unfettered public access and citable persistent identifiers are also acceptable. Sites such as Dropbox that give authors discretion to alter or remove files are not acceptable. Complex databases with an interactive user interface can be made accessible through an institutional website, but an archived static version of the underlying data should be deposited in an appropriate repository. Options for archiving portions of exceptionally large datasets that exceed the capacity of standard repositories should be discussed with the editor prior to publication.
We seek to lower the barriers to data deposition and support making all data available during the pre-publication process. To this end, we have partnered with Dryad, so that data files associated with a submitted manuscript can be deposited or accessed at Dryad directly from our content tracking system. These data files will be available to editors and reviewers during the peer-review process. The Science journals will cover the cost of publishing the data publicly at Dryad if the paper is accepted. Authors who choose to use a different repository should, if possible, provide a corresponding URL when submitting their manuscript, so as to facilitate access by reviewers.
When custom computational methods go beyond simple fitting routines or common algorithms that can be described succinctly in plain English, the underlying code should likewise be deposited in a permanent repository such as Zenodo or Code Ocean. Because GitHub libraries may be modified later by authors in ways that could complicate reproducibility, any code stored on that site should be archived in its current form in a permanent public repository ahead of publication (straightforward instructions for doing so can be found here), and that version should be cited in the paper.
In the interest of tracking data and code reuse and accordingly allocating credit, the Science journals include citations of datasets and archived code with DOIs in the reference list, using DataCite formatting conventions. A consensus guide to citing software, by a working group of multiple publishers, can be found here. These references should be cited in the Data and materials availability section of the acknowledgments, where repository-specific accession codes should also be disclosed.
Field-specific repository recommendations include:
- Molecular and macromolecular structure data. Atomic coordinates and structure factor files from x-ray structural studies or an ensemble of atomic coordinates from NMR structural studies must be deposited and released at the time of publication. Three-dimensional maps derived by electron microscopy and coordinate data derived from these maps must also be deposited. Approved databases are the Worldwide Protein Data Bank [through the Research Collaboratory for Structural Bioinformatics, Macromolecular Structure Database (MSD EMBL-EBI), or Protein Data Bank Japan], BioMag Res Bank, and Electron Microscopy Data Bank (MSD-EBI), and, for synthetic compounds, the Cambridge Crystallographic Data Centre (organic/organometallic) or the Inorganic Crystal Structure Database. We require authors of papers reporting structural data to initiate deposition of the model and data at wwPDB and provide a Preliminary Full or Full validation report from the deposition server (for macromolecules) or CIF and checkCIF files (for synthetic compounds) with their submission. If these are not provided, they will be requested before review. For macromolecular structures, we may also request atomic coordinates and structure factors with map coefficients or electron microscopy density maps during the review process.
- Synthetic organic and medicinal chemistry data. Scanned 1H and 13C NMR spectra may be included in the supplementary materials, but as an alternative we encourage the use of the American Chemical Society’s pilot program to produce zipped files of the full free induction decay datasets for deposition in a general repository.
- DNA and protein sequences. Approved databases are GenBank or other members of the International Nucleotide Sequence Database Collaboration (EMBL or DDBJ) and SWISS-PROT.
- Microarray data. Data should be presented in MIAME-compliant standard format. Approved databases are Gene Expression Omnibus and ArrayExpress.
- Climate, geoscience, and space science. Guidelines on data deposition are provided by the Coalition on Publishing Data in the Earth and Space Sciences (COPDESS), together with a searchable online Repository Finder.
- Materials science data. In addition to general-purpose repositories, authors may consider NOMAD for computational data and the Materials Data Facility for experimental data.
- Ecological data. We recommend deposition of data in Dryad. Our partnership with Dryad is described earlier in this section.
Policies governing materials availability and data and code archiving to support accessibility are detailed in the section on Research standards above. Problems in obtaining access to published data are taken seriously by the Science journals and can be reported at [email protected].
Science, Science Immunology, Science Robotics, Science Signaling, and Science Translational Medicine require that, before manuscript acceptance, authors sign a license agreement granting AAAS rights to use and authorize use of their contribution. The authors retain copyright as well as rights to make certain uses of the work, which are described in detail in the information for authors that accompanies the form. Authors should read this information sheet before signing the agreement. (A separate form is used for authors of papers for which all coauthors are employees of the Australian, Canadian, and/or UK government.)
In general, authors will complete the license form as a “click-through” during the process of uploading the revised manuscript after peer review.
License to Publish: Form [PDF]
Science Advances requires that, before manuscript acceptance, authors sign a license agreement granting AAAS exclusive rights to use and authorize use of their contribution. As an open access journal, Science Advances then asks authors to choose one of two Creative Commons licenses, each of which allows readers to reuse the published content in specific ways. The CC BY-NC license allows readers to distribute, adapt, or reuse articles for non-commercial purposes, while commercial reuse requires permission. The CC BY license allows readers to distribute, adapt, or reuse articles, including for commercial purposes. Further licensing and APC pricing information is found here.
Access Policies for Science, Science Immunology, Science Robotics, Science Signaling, and Science Translational Medicine
Immediately after publication, authors of peer-reviewed research may post the accepted version of their paper on their personal or institutional archival website or in a repository of their choice. The accepted version is the version of the paper accepted for publication after changes resulting from peer review, but before editing by the Science journal copyediting staff, image quality control, and production of the final PDF.
In addition, one author is provided an e-print link, which can be posted on a personal or institutional web page and through which users can freely access the final, published paper on the Science journal’s website.
Peer-reviewed research that has immediate relevance to public health or that reports the reference sequence of a genome, will be made free for reading upon publication Authors may self-archive such papers and distribute their accepted manuscript under a CC BY license. The Science Journals are signatories to the 2016 Statement on Data Sharing in Public Health Emergencies which was updated to address the COVID-19 outbreak. Details can be found here: https://wellcome.org/coronavirus-covid-19/open-data.
Authors of peer-reviewed research submitted to a Science subscription journal on or after 1 January 2023 who are required by their employer or funder to distribute their accepted manuscript under a CC BY license may do so upon publication of the final version by AAAS.
Authors of cOAlition S–funded peer-reviewed research submitted to a Science subscription journal on or after 1 January 2021 may distribute their accepted manuscript under a CC BY license or a CC BY-ND license (if permitted by the funder) upon publication of the final version by AAAS.
Authors from institutions that might limit authors’ ability to grant to AAAS any of the rights described in AAAS’s license must obtain an approved waiver from their institution to publish with the Science journal.
Original research papers are freely accessible with registration on the Science journal’s website 12 months after publication.
Access Policies for Science Advances
Science Advances is an online-only gold open access journal from AAAS. All content is freely available from the date of publication.
Submitted and accepted papers must remain privileged documents and must not be released to the press or the public before publication. The Science journals do allow posting of the submitted version of research papers on preprint servers, but these should not be discussed with the media. We provide embargoed press packages to reporters ahead of publication to facilitate accurate reporting on our published papers. Questions should be referred to the Science Journals Press Package Team ([email protected]).
Science is a member of the Committee on Publication Ethics (COPE), and all Science journals are committed to correcting errors in published papers, and where necessary retracting papers, in order to maintain the integrity of the scientific record. We encourage authors to promptly notify us when there are concerns related to papers that they have published in our journals. These could range from minor corrections to errors or misconduct that affect the core conclusions of the paper. Readers with concerns regarding the integrity of a paper published in a Science journal should contact email@example.com. Concerns will be investigated following COPE guidelines, which give authors a chance to respond. For other substantive and scholarly commentary on articles, readers are encouraged to use our eLetters forum.
For corrections that do not substantively affect any scientific results, the online version and PDF will be corrected, and this will be noted in the article. For corrections that address scientific accuracy, an independent, citable Erratum that transparently explains the corrections will be published online along with the corrected manuscript version(s). A notice of the Erratum will run later in print. Authors who take responsibility for post-publication corrections ensure the integrity of their work. Together we can promote the accuracy of the scientific record and integrity in scientific publications.
In cases of identified errors or irreproducibility of research findings reported in a Science journal paper, a Retraction is likely if the core conclusions are thereby invalidated. An accumulation of errors identified in a paper may cause the editors to lose confidence in the integrity of the data presentation, and the paper may be retracted. In cases of potential large-scale error or misconduct, the author’s institution will be informed, and the journal may publish an Editorial Expression of Concern related to the paper in question while we await the outcome of institutional investigations. Papers will be retracted in cases of research misconduct, in accord with COPE guidelines. The primary goals of Editorial Expressions of Concern and Retractions are transparent notification for our readers and maintaining the integrity of the scientific record. Determinations of scientific misconduct and assigning responsibility to individual authors are the responsibility of the investigating institution. Retraction notices may refer readers to the institution’s investigation reports.