Research Integrity, Plagiarism, Data Fabrication & Image Manipulation Policy

Plant Science Horizons (PSH) is committed to protecting the reliability of the scientific record. Authors, reviewers, editors, and contributors are expected to follow high standards of originality, accurate data reporting, transparent image processing, reproducibility, ethical attribution, and responsible scholarly conduct.

Purpose and scope of this policy

This policy applies to manuscripts submitted to PSH, material under peer review, accepted articles, published articles, supplementary files, images, datasets, code, graphical material, and other content forming part of the scholarly record.

It applies to concerns raised before publication and after publication and covers plagiarism, text recycling, redundant publication, data fabrication, falsification, selective omission, inappropriate image manipulation, misleading presentation, and related forms of research or publication misconduct.

Fabrication, falsification and plagiarism

Forms of plagiarism

Plagiarism may involve text or non-textual scholarly material and may include:

Quotation, paraphrasing and attribution

• Verbatim text from another source must be clearly identified as quoted material and appropriately cited.
• Paraphrasing must genuinely restate the source in the author's own scholarly language and must still cite the original source.
• Citation alone does not justify extensive verbatim reproduction of copyrighted text.
• Sources should be cited at the point where their ideas, methods, data, or interpretations are used.
• Secondary sources should not be cited as though the author personally consulted the primary work unless the primary work was actually examined.

Reuse of an author's own previously published text

Reuse of an author's own writing without transparency or appropriate citation may constitute inappropriate text recycling. Editors assess such overlap according to its extent, location, purpose, originality of the underlying content, and whether the reused material could mislead readers regarding novelty.

Duplicate publication and salami publication

Authors must not submit substantially the same study, dataset, analysis, or conclusions to multiple journals as separate original publications without transparent disclosure and legitimate scholarly justification.

• Closely related manuscripts must be disclosed to the editor at submission.
• Reuse of the same participants, field experiment, samples, sequencing dataset, images, or experimental series across papers must be transparent.
• Dividing one coherent study into minimally differentiated manuscripts primarily to increase publication count is discouraged.
• Secondary analyses of existing datasets may be legitimate when they ask clearly distinct questions and cite the primary dataset/publication appropriately.

Similarity-detection software and editorial interpretation

Submitted manuscripts may be screened using Turnitin or another similarity-detection system. Similarity software identifies textual matches; it does not itself determine whether plagiarism has occurred.

PSH does not apply a single automatic similarity-percentage threshold as the sole basis for editorial action. Editors assess the location, source, concentration, nature, attribution, and scholarly significance of matched material. A low similarity score does not prove originality, and a higher score may contain legitimate methodological or bibliographic overlap.

Fabricated observations, measurements and results

Fabrication includes inventing or reporting observations, experimental replicates, genotypes, field sites, sequencing reads, expression values, analytical measurements, statistical outputs, samples, images, phenotypes, survey responses, or other research information that did not exist.

Authors must be able to demonstrate that reported results derive from genuine research records when reasonable questions arise.

Alteration, omission and misleading presentation of data

Falsification may include:

Outliers, exclusions and missing data

• Data should not be removed simply because they weaken statistical significance or contradict the preferred hypothesis.
• Exclusion criteria should be defined in advance where possible and described transparently.
• Post-hoc exclusions must have a defensible methodological or quality-control basis and should be disclosed.
• Authors should report how missing data were handled when this materially affects analysis.
• Sensitivity analyses should be provided where exclusion decisions substantially affect conclusions.

Raw data, source files and record retention

Authors are expected to retain original research records in accordance with applicable institutional, funder, legal, and disciplinary requirements. Such records may include raw numerical data, laboratory records, instrument files, microscopy images, uncropped blots, sequencing files, metadata, analysis scripts, code, and original figure components.

PSH may request relevant source data during peer review, investigation, or post-publication assessment. Failure to provide reasonable supporting evidence for central results may affect the editorial evaluation of the work.

General principles for scientific images

Images must accurately represent the original observations. Processing is acceptable only when it improves presentation without changing interpretation, selectively altering features, or creating an inaccurate representation of the underlying data.

The same ethical standards apply to photographs, microscopy images, gels, blots, electrophoresis images, colony plates, phenotypic images, fluorescence images, chromatograms, spectra, maps, and other visual scientific records.

Specific expectations for composite and experimental images

• Non-adjacent lanes or images assembled into one figure should be clearly separated by visible boundaries and explained in the legend.
• Controls should be derived from the appropriate experiment or clearly identified when shared across experiments.
• Loading controls, reference controls, and representative images must not be reused deceptively.
• Microscopy panels should use comparable acquisition and processing conditions when direct visual comparison is intended.
• Scale bars should be included where spatial dimensions are scientifically relevant.
• Editors may request uncropped gels, full blots, original microscopy files, metadata, and replicate images.

Panel assembly and reuse

Figure panels may combine material from multiple experiments when scientifically appropriate, provided the assembly is transparent. Boundaries, breaks, or differences in origin must be clearly indicated. A legend should explain the composition when it is not self-evident.

Reuse of an identical image, control, or panel in more than one figure must be scientifically justified and explicitly disclosed.

Integrity of charts, statistics and analytical figures

• Graphs must represent the underlying data accurately and use appropriate axes and scales.
• Truncated axes or other visual choices must not exaggerate effects misleadingly.
• Representative plots should not replace disclosure of variability or independent replication.
• Statistical annotations must correspond to the reported analysis.
• Spectra, chromatograms, sequencing plots, heatmaps, and similar outputs must not be selectively altered to remove inconvenient features without valid disclosed processing.

Integrity of genomic, transcriptomic, metabolomic and computational datasets

• Raw and processed datasets must correspond to the samples and experimental design described in the manuscript.
• Sample labels, accession numbers, metadata, and biological groups must be accurate and internally consistent.
• Filtering, normalization, batch correction, imputation, feature selection, and exclusion criteria must be described transparently.
• Public datasets must not be presented as newly generated data.
• Generated or simulated datasets must be clearly identified as such.
• Code or analysis pipelines must not intentionally suppress or alter outputs to produce misleading conclusions.

Generative AI and synthetic research content

Generative AI must not be used to fabricate experimental observations, synthetic laboratory records, false images, nonexistent citations, invented data, or other material presented as authentic scientific evidence.

Authors remain responsible for verifying the accuracy, originality, provenance, and integrity of all submitted content, including material created or modified with computational or AI-assisted tools.

When and how manuscripts may be screened

PSH may conduct integrity screening before peer review, during review, at revision, before publication, or after publication. Screening may include similarity analysis, examination of figures and metadata, citation checks, data-consistency review, repository verification, or requests for original/source material.

Screening tools assist editorial assessment but do not replace human judgment.

Examples that may trigger further assessment

Editorial assessment of suspected misconduct

Opportunity to clarify concerns

Unless immediate action is necessary to protect the scholarly record, authors will normally be given an opportunity to respond to substantive concerns and provide relevant documentation. Responses should address the specific concern rather than rely only on general assurances.

Where multiple authors are involved, the journal may communicate with the corresponding author, all authors, or selected authors depending on the issue.

Confidential handling of integrity concerns

Suspected misconduct will be handled as confidentially as reasonably possible. Information will be shared only with individuals or organizations whose involvement is necessary to assess, investigate, resolve, or document the concern. Allegations should not be treated as established facts before appropriate assessment.

Role of institutions and research-integrity bodies

Journals evaluate the reliability and publication status of scholarly content; institutions and authorized research-integrity bodies may have greater authority and access to determine whether formal research misconduct occurred.

When serious concerns cannot be resolved through journal-level assessment, PSH may contact the relevant institution, employer, funder, ethics body, or other appropriate authority and may consider information supplied by a formal investigation.

Possible actions for a submitted manuscript

Post-publication integrity concerns

Publication does not end the journal's responsibility to maintain the scholarly record. Credible post-publication concerns will be assessed whether raised by authors, readers, reviewers, editors, institutions, integrity specialists, or other identifiable sources.

The journal will focus on the evidence and reliability of the published record rather than the status or identity of the person raising the concern.

Retraction is intended to correct the literature

Retraction is not primarily a punishment. Its purpose is to alert readers when a publication or substantial part of it cannot be relied upon. Retraction notices should identify the affected article and explain the reason sufficiently clearly to protect the scholarly record while avoiding unsupported allegations.

Editorial action when authors do not agree

Necessary corrections, expressions of concern, or retractions may proceed when justified by the evidence even if one or more authors disagree. The journal will accurately state the basis for the action and, where appropriate, indicate which authors agree or disagree.

Concerns raised by readers and third parties

Reports should identify the article or manuscript and describe the concern with sufficient specificity for assessment. Supporting evidence, figure locations, data inconsistencies, or source comparisons are helpful.

PSH will not dismiss a credible concern solely because it is critical of the journal or authors. At the same time, repetitive, abusive, defamatory, or unsupported allegations may not warrant continued editorial engagement.

Good-faith concerns and retaliation

PSH expects concerns to be raised in good faith and handled professionally. The journal does not support retaliation against individuals for responsibly raising credible research-integrity concerns. Deliberately false or malicious accusations are themselves inconsistent with responsible scholarly conduct.

Misconduct detected during peer review

Reviewers who identify possible plagiarism, duplicated images, fabricated data, manipulated results, or other integrity concerns should report the issue confidentially to the editor with enough detail to permit assessment.

Reviewers must not independently contact the authors, publicly accuse them, or use confidential manuscript material for personal advantage.

Editors' responsibilities in integrity cases

• Take credible concerns seriously and avoid premature conclusions.
• Maintain appropriate confidentiality and documentation.
• Separate conflicts of interest from the investigation and recuse conflicted editors.
• Seek relevant evidence, including raw data or institutional input when necessary.
• Use proportionate editorial actions based on the reliability of the scholarly record.
• Correct the record promptly when the available evidence supports action.

Responsibility of all authors

All authors share responsibility for the integrity of the published work within the scope of their contribution and should cooperate with legitimate requests to clarify concerns. The corresponding author should facilitate communication but does not replace the responsibility of individual coauthors.

Authors should promptly notify the journal if they discover a significant error, duplicated material, incorrect image, data problem, or other issue that could affect the reliability of a submitted or published article.

Editing services, laboratories and external providers

Authors remain responsible for the integrity of material generated, edited, analyzed, or prepared by third-party laboratories, sequencing providers, statistical consultants, language-editing services, figure-preparation services, or other contractors. Outsourcing does not transfer responsibility for accuracy or ethical compliance.

Reused figures and third-party material

Even when reuse is properly cited, authors must obtain permission when required by copyright or license terms. Modified or adapted figures should be identified appropriately. Citation does not replace permission where permission is legally required.

Errors that do not constitute misconduct

Not every error is research misconduct. Transcription mistakes, incorrect labels, coding errors, statistical errors, or other unintended problems may occur despite good-faith research practice.

The appropriate response depends on the effect of the error on the published conclusions. Material errors may require correction or retraction even when no misconduct occurred.

Requests for reconsideration

Authors may request reconsideration of an integrity-related editorial decision when they can provide substantive new evidence, identify a material factual error in the assessment, or demonstrate a significant procedural concern. Reconsideration does not guarantee reversal of the original decision.

Editorial guidance and proportional response

PSH aims to address integrity concerns using evidence-based, proportionate, and transparent editorial procedures informed by recognized publication-ethics guidance, including relevant guidance from the Committee on Publication Ethics (COPE).

The journal may update its procedures as publication standards, technologies, and best practices evolve.

Before submission, authors should confirm

Commitment to Research Integrity

Plant Science Horizons is committed to preserving the accuracy, credibility, transparency, and trustworthiness of the scientific record. Concerns involving plagiarism, data fabrication, falsification, inappropriate image manipulation, redundant publication, or related integrity problems will be evaluated carefully and addressed in proportion to the available evidence and their effect on the reliability of the work.