Clinical research
Trial design, endpoints, protocol alignment, missing outcomes and the reporting of clinical and observational studies.
Independent scientific review
Put your paper
to the test.
An early technical review can save months of back-and-forth in peer review.
We review the statistics and code behind life sciences manuscripts, check every claim against the evidence, and edit the paper for clarity.
RED TEAM REVIEWPLATE / 01
FIG 01 / CLAIM ↔ EVIDENCEILLUSTRATIVE
Built for research teamsUniversitiesAcademic medical centresClinical research groupsLife sciences R&D
01 / Scope of review
What we scrutinise
We bring expertise in scientific communication and peer review to work ranging from clinical trials to computational biology. We review the paper alongside its figures, supplements and analysis code.
Statistical review
We identify statistical issues in model choice, assumptions, missing data, multiplicity and uncertainty, then explain how they could affect the conclusions.
Code review & pull requests
We inspect research codebases, trace reported results to the code that produced them, and can submit pull requests with proposed fixes directly to your repository.
Computational biology
We examine omics pipelines, model validation, data leakage, batch effects and whether the methods are described well enough to reproduce.
Claim-by-claim review
We evaluate every claim in the paper, from abstract to figure captions, against the underlying analyses or cited literature. We flag claims that need stronger support or more precise wording.
Language & readability
We edit for clarity, flow and precise scientific language so readers can follow the argument and understand what the results show.
02 / How we harness frontier AI
Expert-led review, powered by AI
We deploy bespoke agents built on advanced AI models. For complex studies, they can run thousands of checks across data, figures and code. Automating these checks helps keep your review costs down.
- DATA
Check the inputs
Compare cohort counts across analysis files and tables. Flag missing values, duplicate records and unexpected exclusions.
- FIGURES
Check what’s shown
Compare plotted values with analysis outputs. Check labels, units and whether captions describe the right analysis.
- CODE
Trace the results
Follow reported results back to the scripts that generated them. Flag differences in filtering, model settings or analysis inputs.
An expert leads every review. Your reviewer sets the scope, directs the agents and checks the findings that could change your conclusions or require substantial revisions before delivering the report.
03 / Sample reviews
From critique to revision
Two independent reviews of public research. Each assesses the paper’s contribution, prioritises revisions and shows the checks behind the findings.
Clinical research · Published article
COV009 vaccine follow-up
Reconciles the article, protocol and registry. Rebuilds rare-event comparisons from the appendix and examines how incomplete follow-up affects interpretation.
Read the COV009 reviewGenomics · Preprint
DiSCoVeR STEC genomics
Joins isolate-level supplementary tables to check clinical outcomes, traces a figure discrepancy to the underlying counts and tests sample-ID handling in the linked pipeline.
Read the DiSCoVeR review04 / Large studies
50 supplementary figures?
We review the whole study.
No supplement is too big. We review large datasets, multi-cohort analyses and extensive figure sets, tracing results across the paper, supplement and code.
Supplementary figure index50 figures
- S08 Cohort counts
- Reconcile the sample sizes in the cohort diagram, analysis tables and main text.
- S27 Sensitivity analysis
- Check whether the result holds under a different exclusion rule and whether the abstract reflects the change.
- S46 Figure provenance
- Trace the plotted values to the generating script and verify that the caption describes the right analysis.
05 / Process
How a review works
Start with one manuscript or discuss a recurring review process for your research group.
- 01
Agree the scope
Tell us the study type, target journal if selected, and submission date. We agree the materials, technical priorities and timeline.
- 02
Review the evidence
Your reviewer directs AI agents to check the manuscript claim by claim and examine the agreed analyses, code and supporting materials.
- 03
Prioritise revisions
You receive a written review with ranked findings and suggested actions. Depending on scope, we can also provide tracked manuscript edits and submit pull requests.
06 / Enquiries
Discuss your
next submission.
We welcome enquiries from research groups in Saudi Arabia and internationally. A short outline of your study, its current stage and your submission window is enough to get started.
Please keep manuscript files and patient data out of your initial enquiry. We can agree how to share review materials once we’ve scoped the project.
[email protected]