In one sentence. SGSJ is run by its editors, answers to its readers and authors, belongs to no university or ministry, and publishes nothing it cannot source.
1. Who runs the journal
SGSJ is published by the SudaGene Consortium through Gene Academy. Editorial decisions are made by the Founding Editor and the section editors, and by nobody else. The publisher does not approve, veto or review individual manuscripts.
Each of the four sections has its own editor and reviewer pool. A manuscript that crosses sections is handled jointly by the relevant editors.
2. Editorial independence
- SGSJ is not an organ of any university, ministry, council, professional body or existing journal. It works with all of them by agreement, and on equal terms.
- No funder, partner or institution has any right of review over a manuscript, and none has ever been offered one.
- Editors and reviewers serve in a personal scholarly capacity. They do not represent their employers, and their employers do not speak for the journal.
- An author's stated affiliation records where that person works. It is not an endorsement of the paper by that institution.
- The journal's voice is academic. The personal views, teaching and public outreach of its founder and members are their own, are labelled as such, and play no part in editorial decisions.
3. Conflicts of interest and recusal
Three rules, each absolute.
- No editor handles their own work. Any manuscript an editor has written or co-written is assigned to a different editor, and the editorship is declared on the published paper.
- No credential-checker edits the person they checked. Any platform lead or administrator, including the Founding Editor, who performs or signs off a member's credential verification is recused from editing or reviewing any manuscript that member submits or co-authors.
- Competing interests are declared, not weighed privately. Authors, editors and reviewers declare financial, institutional and personal interests. Declarations are published with the paper.
Where a conflict cannot be avoided because the field is small, it is disclosed in the published record rather than concealed.
4. Peer review
- Editorial screening within two weeks of submission.
- At least two independent reviewers per research article, double-anonymised.
- Working papers are published before review and are labelled as such on every page and in every citation.
- Reviewers are asked to judge the work, not the author's institution, country or language.
- Reviewer reports are confidential. A reviewer may ask to be named on publication.
5. Authorship and credit
Authorship follows the ICMJE criteria. Contribution is recorded using the CRediT taxonomy, for every author and every named contributor.
Field collectors, clinicians, translators, interviewers and data extractors are named. Where their contribution does not meet the authorship criteria, they appear in the acknowledgements by name, with their role stated.
Nobody is added to an author list as a courtesy, and nobody who meets the criteria is left off.
6. Consent to be listed
No person's name appears on this site — as an editor, a reviewer, a verified member or a contributor — without their explicit recorded agreement. Consent is recorded with a date and may be withdrawn at any time, after which the name is removed.
This applies to editorial appointments. An invitation is not an appointment, and a name is published only after the person has accepted in writing.
7. Use of AI and computational tools
- Tools used by the editorial office, for screening, translation support or in-silico analysis, are named publicly, tested adversarially before use, and audited. Criticism of them is welcome and published.
- AI systems are never authors. Authors state any AI assistance in their methods section.
- Every AI-assisted analysis is checked by a named person who takes responsibility for it.
- Results from computational modelling are labelled in-silico and treated as hypothesis-generating until confirmed by empirical work.
- Any submission that used an AI tool at any stage carries a verification log: one row per factual claim that originated from or was checked by a tool, with the source checked against and the outcome. A claim recorded as unverified is removed or explicitly hedged; it cannot stand as a flat assertion.
8. Corrections, retractions and the published record
Published work is corrected openly, never quietly amended.
| Situation | What happens |
|---|
| Minor error | A dated correction notice is attached to the article and linked from it. |
| Error affecting the conclusions | A correction notice, and the abstract is amended to state what changed. |
| Unreliable findings, honest error or misconduct | Retraction, following COPE guidance. The article stays online, marked retracted, with the reason stated. |
| Dispute over authorship or data | An expression of concern while the matter is investigated. |
Nothing is removed from the site once published, except where a legal obligation or a serious risk to a person requires it. In that case a notice records that something was removed and why.
9. Complaints and appeals
An author may appeal a decision by writing to the editorial office with the grounds. The appeal is handled by an editor who was not involved in the original decision. If no such editor exists, an external reviewer is asked.
Complaints about an editor, including the Founding Editor, go to the editorial office and are referred to a section editor who is not the subject of the complaint.
10. Data, ethics and consent in research
- Research involving people requires ethics approval from a committee in the country where the work was done, named in the manuscript.
- How consent was obtained must be described, including the modality. "Informed consent was obtained" without more is not sufficient where literacy, displacement or authority relationships are in play.
- Research conducted in Sudan is expected to include Sudanese investigators as authors, not only as hosts or facilitators.
- Data are deposited openly where ethics and consent allow, with the deposit cited in the paper.
11. Open access and licensing
Every article is free to read on publication and free to publish. There are no article-processing charges, submission fees or page charges. Articles are published under CC BY 4.0 and authors retain copyright.
12. Standards this policy is written against
COPE Core Practices, the ICMJE recommendations on authorship and disclosure, the CRediT contributor taxonomy, the FAIR data principles, and the DOAJ criteria for open-access journals. Where SGSJ holds a formal membership or listing it says so; where an application is pending or planned, it says that instead.