Data Integrity and Reproducibility

1. Purpose and Scope

AMG Transcend Association requires research published in its journals to be based on data and other evidence that are generated, processed, analyzed, presented, and reported honestly and transparently.

This policy establishes publisher-level principles concerning data integrity, methodological transparency, reproducibility, statistical and computational reporting, availability of underlying data, data retention, and the assessment of concerns affecting the reliability of research findings.

The requirements apply as relevant to the nature of the research and may be supplemented by journal-specific instructions, disciplinary standards, and recognized reporting guidelines.

2. Data Integrity

Authors must ensure that data and other research records supporting a manuscript are authentic, accurate, appropriately represented, and consistent with the reported findings.

Authors must not:

• fabricate data, observations, measurements, results, or research records;
• falsify or inappropriately alter data or results;
• selectively omit data for the purpose of creating a misleading interpretation;
• manipulate analyses, figures, images, tables, or graphical presentations in a manner that misrepresents the underlying evidence;
• misrepresent simulated, synthetic, processed, estimated, or derived data as directly observed or experimentally obtained data;
• conceal material problems affecting the reliability or provenance of a dataset;
• report results in a manner materially inconsistent with the underlying evidence.

Legitimate data cleaning, processing, transformation, exclusion, normalization, correction, or other analytical procedures are permitted where scientifically justified and appropriately documented.

Such procedures must not be used to create a misleading representation of the underlying evidence.

3. Raw, Processed, and Derived Data

Authors should maintain a clear distinction, where relevant, between:

• original or raw data;
• cleaned or processed data;
• transformed or normalized data;
• derived variables or calculated results;
• final data presented in figures, tables, statistical analyses, or other outputs.

Processing steps must be sufficiently documented to permit the relationship between the underlying data and the reported results to be understood and evaluated.

Data processing must not remove, obscure, introduce, or alter scientifically relevant information in a misleading manner.

Where exclusions, corrections, substitutions, transformations, imputations, or other material changes are made to the data, the scientific basis for those actions must be documented and reported where necessary for appropriate interpretation of the study.

4. Traceability and Research Records

Research must be documented in a manner that supports reasonable traceability from reported findings to the relevant underlying observations, measurements, data, analyses, or research records.

Depending on the type of study, relevant records may include:

• original or underlying datasets;
• laboratory or experimental records;
• instrument outputs;
• processed datasets;
• analytical workflows;
• statistical files;
• code or scripts;
• image or figure source files;
• protocols;
• metadata;
• database identifiers;
• computational parameters;
• other documentation necessary to understand how reported results were obtained.

Not every type of research will generate all of these materials.

Documentation expectations must be proportionate to the study design, discipline, methods, and nature of the reported findings.

5. Methodological Transparency and Reproducibility

Authors must describe methods with sufficient clarity and detail to permit appropriate scientific evaluation and, where reasonably possible, reproduction, replication, or independent verification of the research approach.

Relevant information may include:

• study and experimental design;
• materials, samples, datasets, and data sources;
• experimental procedures;
• measurement and analytical methods;
• inclusion and exclusion criteria;
• data preprocessing and transformation;
• statistical methods;
• key instruments and equipment;
• software, algorithms, packages, and relevant software versions;
• computational parameters and settings;
• database versions or accession information;
• modifications to established methods;
• other information necessary to understand how the study was conducted and analyzed.

Previously published methods may be cited where appropriate, but material modifications and study-specific procedures must be described sufficiently.

Claims concerning reproducibility or replicability must not imply that identical results are guaranteed under all circumstances.

The objective is to provide sufficient methodological transparency for independent researchers to evaluate, reproduce, replicate, verify, or extend the work as reasonably appropriate to the research design.

6. Statistical and Analytical Integrity

Statistical and analytical methods must be appropriate to the study design, data characteristics, research question, and nature of the conclusions being drawn.

Authors must report relevant analytical procedures with sufficient detail to permit appropriate evaluation.

Where applicable, manuscripts should transparently describe matters such as:

• statistical tests and models;
• assumptions relevant to the analyses;
• sample size and analytical population;
• sample-size justification or power considerations where relevant;
• treatment of missing data;
• exclusions and outliers;
• transformations;
• multiple comparisons or adjustments;
• measures of variability or uncertainty;
• relevant effect estimates and confidence intervals;
• prespecified and exploratory analyses;
• other analytical decisions materially affecting interpretation.

Authors must not selectively choose, modify, repeat, suppress, or report analyses primarily to obtain a preferred, favorable, or statistically significant result.

Exploratory or post hoc analyses must not be presented as though they were prospectively specified where that distinction is material to interpretation.

7. Computational Research, Code, and Software

Research that substantially depends on computational methods must provide sufficient information to permit the analytical process to be understood and critically evaluated.

Where applicable, authors should identify:

• software and relevant versions;
• computational tools, libraries, or packages;
• algorithms or models;
• important parameters and settings;
• databases and relevant versions or access information;
• preprocessing and analytical workflows;
• custom code or scripts relevant to the reported results.

Authors are encouraged to make analytical code and scripts publicly available where legally, ethically, technically, contractually, and practically possible.

Where code cannot be made public, the manuscript must still provide sufficient methodological information to permit meaningful evaluation of the computational approach.

Proprietary software, commercial restrictions, or limitations on code availability must not be used to conceal methodological information necessary for evaluating the validity or reliability of the reported analysis.

8. Images, Figures, and Other Data Representations

Images, spectra, chromatograms, gels, blots, microscopy data, graphical outputs, and other visual representations of research data must accurately reflect the underlying evidence.

Adjustments or processing must not selectively enhance, remove, obscure, move, introduce, duplicate, or alter features in a manner that could mislead readers about the underlying data.

Where image, signal, or other data processing is scientifically necessary, relevant procedures should be applied consistently and described where required for interpretation.

Figures and tables must be internally consistent with the manuscript and with the underlying data from which they were derived.

Detailed requirements concerning acceptable and unacceptable image processing and assessment of suspected manipulation are addressed in the Image Manipulation and Publication Ethics and Research Integrity policies.

9. Reuse and Use of Third-Party Data

Previously published, publicly available, licensed, proprietary, repository-based, or other third-party data must be clearly identified and appropriately attributed.

Authors are responsible for ensuring that their use of third-party data complies with applicable:

• licenses;
• permissions;
• consent requirements;
• data-use agreements;
• ethical approvals;
• confidentiality obligations;
• intellectual-property requirements;
• legal or regulatory requirements.

Previously existing data must not be presented as newly generated data.

The provenance of important external datasets and any material processing, transformation, selection, or integration performed by the authors must be reported sufficiently to support evaluation of the analysis.

10. Data Availability Statements

AMG Transcend Association journals require transparent information concerning the availability of data underlying research findings.

Articles reporting original research or original analyses of data must include a Data Availability Statement.

The statement must indicate, as applicable, whether supporting data are:

• available in a public repository;
• included in the article or supplementary materials;
• available from the authors upon reasonable request;
• available through controlled or restricted access;
• obtained from an identified third-party source;
• subject to ethical, legal, contractual, privacy, confidentiality, commercial, intellectual-property, or other justified restrictions;
• not applicable because no new data were created or analyzed.

Where data are deposited in a repository, persistent identifiers, accession numbers, repository links, or other information necessary to locate the data should be provided where available.

Data Availability Statements must accurately describe the actual accessibility of supporting data and must not state or imply that data are available where meaningful access cannot reasonably be provided.

Where data cannot be shared, the reason for the restriction should be stated with sufficient clarity to permit readers and editors to understand the basis for the limitation.

11. Responsible Data Sharing

AMG Transcend Association supports responsible sharing of research data where this can be achieved without violating ethical, legal, privacy, confidentiality, intellectual-property, contractual, security, or other legitimate restrictions.

Public sharing of all research data is not required in every circumstance.

Where unrestricted public sharing is inappropriate, responsible alternatives may include:

• anonymized or de-identified datasets;
• controlled-access repositories;
• restricted data access;
• appropriate data-use agreements;
• sharing of non-identifiable subsets;
• provision of aggregated information;
• another access mechanism consistent with applicable requirements.

Restrictions on data sharing must be genuine and should be described transparently in the Data Availability Statement.

Authors must not cite confidentiality, commercial sensitivity, privacy, intellectual property, or another restriction as a pretext for withholding data where that restriction does not genuinely apply.

12. Sensitive, Personal, and Confidential Data

Data-sharing and reproducibility expectations do not override obligations to protect research participants, confidential information, legally protected data, proprietary information, security-sensitive information, or other restricted material.

Human-participant data must be handled consistently with applicable informed consent, ethics approval, privacy, confidentiality, institutional, and data-protection requirements.

Authors must not publicly disclose identifiable or sensitive information merely to satisfy a data-sharing expectation.

Where access is restricted, authors should provide sufficient information concerning the nature and reason for the restriction to allow readers and editors to understand the basis for limited availability.

13. Data Retention

Authors must retain research data and relevant supporting records in accordance with applicable legal, regulatory, institutional, contractual, disciplinary, sponsor, and funder requirements.

AMG Transcend Association does not impose a single universal data-retention period across all research disciplines and study types.

Data and documentation supporting published findings should be retained for a period sufficient to permit reasonable verification of the work and assessment of legitimate post-publication concerns, subject to applicable requirements and practical, ethical, or legal limitations.

Once authors become aware of a credible concern, formal inquiry, or ongoing assessment affecting particular research records, relevant materials must not be intentionally destroyed, altered, concealed, or made unavailable in order to prevent appropriate examination of the issue.

14. Requests for Underlying Data or Documentation

Editors or reviewers may request access to relevant underlying data, source materials, analytical documentation, original files, code, or other supporting information where reasonably necessary to assess:

• the validity or reliability of reported findings;
• inconsistencies within the manuscript;
• methodological or statistical concerns;
• image or data-integrity concerns;
• reproducibility or traceability;
• compliance with applicable ethical requirements;
• a credible post-publication concern.

Requests must be proportionate to the matter under assessment and must take account of legitimate ethical, legal, confidentiality, privacy, contractual, intellectual-property, security, and data-protection restrictions.

Where unrestricted transfer or disclosure is not possible, editors may consider alternative mechanisms for independent assessment where appropriate.

Failure to provide relevant data or documentation without an adequate explanation may affect the journal’s ability to establish the reliability of the work and may result in further editorial assessment or action.

15. Responsibilities of Authors

Authors are responsible for ensuring that:

• data are reported honestly and accurately;
• analytical, processing, and data-management procedures are scientifically justified;
• methods and analyses are described transparently;
• material exclusions, transformations, corrections, substitutions, or deviations are appropriately reported;
• data availability information accurately reflects actual access conditions;
• relevant research records are retained in accordance with applicable requirements;
• legitimate editorial requests for supporting information receive appropriate cooperation;
• significant errors affecting submitted or published findings are promptly reported to the journal.

Responsibility for specific datasets, experiments, analyses, code, or other research components may differ among authors according to their contributions.

All authors must nevertheless cooperate in resolving legitimate questions concerning the integrity, accuracy, or reliability of the submitted or published work.

16. Responsibilities of Editors and Reviewers

Editors and reviewers should consider, where relevant to the nature of the manuscript, the consistency between research questions, study design, methods, data, analyses, results, and conclusions.

Editors may request clarification, additional methodological information, statistical review, underlying data, source files, code, original images, or other evidence where a specific concern requires further examination.

Reviewers should confidentially report suspected data fabrication, falsification, inappropriate manipulation, unexplained material inconsistencies, or other significant integrity concerns to the responsible editor.

Neither editors nor reviewers should treat an unusual result, analytical disagreement, missing methodological detail, inability to reproduce a finding, or unexpected outcome as automatic evidence of misconduct.

17. Data Integrity and Reproducibility Concerns

Concerns affecting data integrity, methodological transparency, or reproducibility must be assessed according to their nature, available evidence, scope, and effect on the reliability of the work.

Assessment may involve:

• clarification from the authors;
• examination of underlying or source data;
• review of original files, metadata, or analytical records;
• additional statistical, methodological, image, computational, or other specialist assessment;
• comparison between reported results and supporting data;
• examination of related publications where relevant;
• referral to an institution, funder, regulator, or other appropriate body where formal investigative authority is required.

Honest error, methodological disagreement, inadequate reporting, irreproducibility, and deliberate fabrication or falsification must not be treated as equivalent.

The purpose of editorial assessment is to determine whether the submitted or published work remains sufficiently reliable and whether action is necessary to protect or correct the scholarly record.

18. Post-Publication Action

Authors must cooperate with legitimate assessments of data-integrity, methodological, or reproducibility concerns identified after publication.

Where an error is confirmed but the principal findings remain reliable, correction or clarification may be appropriate.

Where substantial uncertainty remains concerning the reliability of published findings, an expression of concern may be appropriate while further assessment is undertaken.

Where fabrication, falsification, serious data manipulation, or another problem invalidates or substantially undermines the reliability of the work, retraction or other appropriate action may be required.

The absence, unavailability, or loss of underlying data does not automatically establish misconduct.

However, inability to verify essential supporting evidence may affect whether the journal can continue to regard the published findings as sufficiently reliable.

Post-publication actions must be proportionate and conducted in accordance with the Publication Ethics and Research Integrity and Corrections, Retractions and Expressions of Concern policies.

19. Relationship with Other Policies

This policy should be read together with applicable AMG Transcend Association and journal policies concerning:

• Publication Ethics and Research Integrity
• Research Ethics
• Human and Animal Research Ethics
• Informed Consent and Clinical Trials
• Authorship and Contributor Responsibilities
• Conflict of Interest and Competing Interests
• AI and Generative Tools
• Image Manipulation
• Peer Review Principles
• Corrections, Retractions and Expressions of Concern
• Complaints and Appeals

Individual journals may establish additional data, repository, reporting, statistical, computational, reproducibility, or documentation requirements appropriate to their disciplines, methodologies, and article types.