Corpus 1,272 assessed · 1,173 scored · 643 reproduced ≥75 · 168 flagged ·∅ 74.1/100
← New search

Monocytes serve as Shiga toxin carriers during the development of hemolytic uremic syndrome.

Cell Mol Biol Lett · 2025
L1 84/100 3/4
Why this verdict

The main results reproduced: recomputed values matched the published ones within tolerance.

Reproduced on the brainbox compute brainarbeit.com
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Concordant (toward reproduced)
All content-critical questions reproduced
-4 pts
From: Q7 · Core claim 🟢
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score -5
✓ What held up
  • Same input data as the authors
  • Reported values were directly comparable
  • No authors-side cause for any deviation
  • Reported values are derivable from the shared data
  • Any deviation was negligible
  • The central claim held under reproduction
  • Overall, the reproduction was clean
What did not (or only partly)
  • 🟡A deviation arose in the data or preprocessing
How its reproducibility compares
84/100
Reproducibility score
0.6 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 63% of all assessed papers rank 392 of 1173 scored

A 0–100 reproducibility-quality score from the per-question grades, shown as a z-score: standard deviations above (+) or below (−) the mean of comparable assessments.

Reproduction agent’s raw note

Described well enough and reproduced ~1:1. The paper's own scRNA data is GEO GSE252352/GSM8000021 (the brief's GSE202109 is an unrelated cited reference atlas; we used the correct accession). The repo (Xinlei672/Stx2-cell) is a loose authors' Seurat script with local-only inputs; we re-implemented its documented steps and ran them on the shipped 10x matrix on «our HPC» (SLURM 2175624, Seurat 4.3.0.1, Matrix downgraded to 1.6.1.1 to fix a validObject crash). EXACT: 9410 cell barcodes (C1, byte-exact from the raw matrix); the four cell lineages Monocyte/Neutrophil/NKT/Mast (C5); 13 clusters at the repo's res 0.6 (matches its 0-12 cluster rename). The paper's CENTRAL thesis -- monocytes 70% of Stx2-bound cells -- reproduced to 70.47%, with neutrophils 27.49% (<30%), via objective marker-driven cluster annotation. Only C2 is partial: post-QC 9276 vs reported 8681 -- the 8681 additionally needs the DoubletFinder doublet-removal step (7.5% rate; 9276*0.925=8580, consistent), which would not install against the Seurat-v4 stack (HEAD requires SeuratObject 5). NOT attempted (out of scope): Stx2-B-FITC flow-cytometry binding assays, western blots, in-vivo mouse HUS, and the ccl_heatmap script (its chemokine table is not shipped). No fabrication concern: every reproduced value is directly derivable from the shipped data+code.

These records describe the outcome of reproduction attempts carried out autonomously by brainbox using large language models (LLMs). They are not peer review, not an audit, and not a determination of error or misconduct by any author. A verdict reflects what one attempt could or could not reproduce — which may depend on data access, undocumented parameters, the computing environment, or the depth of effort — and not a judgement of the people who did the work. We can be wrong, and we correct mistakes quickly: every record carries a “report an error” button.

Assessment versions

Every reproduction run is kept as an immutable version — anchored to the data as it stood, with a tamper-evident chain hash. A rerun (e.g. after an author updates a deposit) adds a new version; the previous one stays on record.

  1. v1 current initial assessment Score 84
    assessed: 2026-06-14 ⛓ 7e26211853ef
✎ I am an author of this paper

Updated or fixed a deposit, or is there an erratum? Ask us to re-run the metrics. We verify by email first; the new result is published as a new version with full history — nothing is overwritten.

Reason for the rerun

We email you a confirmation link first. The rerun is an objective re-measurement — it cannot change the verdict in your favour, only ask us to look again.

Provenance — full disclosure

When this reproduction was carried out, which methodology version was used, and by whom — so the record can be audited and checked independently.

Reproduced
2026-06-14
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-15
no human curator yet
Last updated
2026-08-05

Provisional, curator- or AI-assessed, and independently checkable. A reproduction outcome states what one attempt could reproduce — not a judgement of the authors.

Deep full-text extraction

Model: opus
Founding hypothesis

The paper tests which circulating cell type carries Shiga toxin (Stx2) from STEC-infected intestines to the kidney during hemolytic uremic syndrome (HUS), hypothesizing that monocytes serve as the primary Stx2 carriers.

Core claims
  • Peripheral CD11b+CD14+ monocytes are the primary carriers that transport Stx2 to the renal region, with nearly all monocytes binding Stx2-B versus <10% of neutrophils at lower affinity finding
  • Monocytes bind Stx2-B largely through Toll-like receptor 4 (TLR4) mechanism
  • Stx-laden monocytes transport Stx2 to human renal glomerular endothelial cells (HRGEC) and induce HRGEC apoptosis finding
  • In EDL933 infection-induced HUS mice, Stx2-positive monocytes are present in peripheral blood and infiltrated kidney tissues finding
  • Depleting monocytes with a CD14 neutralizing antibody or blocking CCL2-mediated chemotaxis with bindarit mitigates LPS/Stx2-induced kidney injury and dysfunction finding
  • Single-cell sequencing of Stx2-B-bound peripheral white blood cells identifies the toxin-binding cell populations method
  • Recombinant Stx2-B was purified from STEC EDL933 and chemically conjugated to FITC as a binding probe method
  • CD14 neutralizing antibody and CCL2 inhibition (bindarit) represent promising therapeutic approaches for Stx-induced HUS resource
Experimental setups
Assay System Perturbation Readout Platform
single-cell RNA-seq (scRNA-seq) Stx2-B-FITC-positive human peripheral white blood cells (FACS-sorted) Stx2-B-FITC incubation single-cell transcriptomes / cluster identity of Stx2-binding cells 10x Chromium Single Cell 3' Reagent Kit v2; Illumina HiSeq4000; Cell Ranger v1.2.0; Seurat v3.0
flow cytometry toxin-cell binding assay human peripheral white blood cells; THP-1 cells; HRGECs Stx2-B-FITC (100 ng/ml) or Stx2-CY5 (2 ng/ml) treatment Stx2 binding, CD14/CD11b co-staining, binding affinity over time BD FACS Aria SORP; Beckman flow cytometer
toxin delivery / co-culture cytotoxicity assay CFSE-labeled THP-1 monocytes co-cultured with HRGECs Stx2 (2 ng/ml) pre-bound to THP-1 HRGEC apoptosis and cytotoxicity index Keygen apoptosis kit KGA1030; CCK-8 (Dojindo); SpectraMax (450 nm)
extracellular vesicle (EV)-toxin binding assay human monocytes (CD11b+CD14+) and neutrophils (CD11b+CD14-) derived EVs incubated with HRGECs Stx2-B-FITC (100 ng/ml) or Stx2 (2 ng/ml) HRGEC apoptosis WGA magnetic beads; Keygen apoptosis kit KGA1030
TLR4 neutralization / binding assay & in silico docking THP-1 cells; TLR4 (O00206) and Stx2-B (P09386) sequences TLR4 neutralizing antibody (10 μg/ml) vs IgG2a control Stx2-B binding inhibition; predicted docking interface BioLegend antibodies; AlphaFold Server; PyMOL; Leica immunofluorescence
EDL933 infection-induced HUS mouse model + flow cytometry/immunofluorescence CD-1 mice infected with E. coli O157:H7 EDL933 (10^10 CFU); kidney tissue streptomycin + EDL933 oral infection Stx2-positive monocytes in blood and kidney anti-mouse CD14 (BioLegend 150105); Leica confocal
LPS/Stx2-induced HUS mouse model + intervention male C57BL/6 mice LPS (300 μg/kg) + Stx2 (500 ng/kg) ± CD14 neutralizing antibody (50 μg/day) or bindarit (100 mg/kg/day) plasma creatinine, kidney injury/histopathology, survival Creatinine Assay Kit (Sigma MAK080); PAS/H&E staining; Olympus microscope
Western blot renal tissue / cells LPS/Stx2 treatment Bax and Cleaved Caspase-3 apoptosis markers, normalized to β-actin CST antibodies; ImageJ
Key results
  • Nearly all monocytes bound Stx2-B with strong affinity while less than 10% of neutrophils bound, at lower affinity ~all monocytes vs <10% neutrophils
  • Monocytes occupied nearly 70% of all Stx2-B-bound peripheral blood cells by scRNA-seq, while less than 30% were neutrophils ~70% monocytes vs <30% neutrophils
  • Stx2-B-bound cells were mainly monocytes across 8681 quality-filtered single-cell transcriptomes
  • TLR4 neutralizing antibody reduced Stx2-B binding to THP-1 cells, supporting TLR4 as the binding receptor
  • Stx2-loaded monocytes co-cultured with HRGECs induced HRGEC apoptosis/cytotoxicity
  • Stx2-positive monocytes detected in peripheral blood and infiltrated kidney tissues of EDL933-infected mice
  • CD14 neutralizing antibody or bindarit (CCL2 inhibition) significantly alleviated kidney injury and dysfunction in LPS/Stx2-treated mice
Key statistics
  • count 8681 quality-filtered single cells analyzed (from 9410 reported cell barcodes) (scRNA-seq of Stx2-B-positive cells)
  • count median 2437 genes per cell; mean 45,584 reads per cell; 97.0% sequencing saturation (Cell Ranger QC metrics for Stx2-B-positive sample)
  • count nearly 70% of Stx2-B-bound peripheral cells were monocytes (proportion of monocytes among toxin-bound cells)
  • count less than 30% of Stx2-B-bound cells were neutrophils (neutrophil proportion among toxin-bound cells)
  • count less than 10% of neutrophils associated with Stx2-B (neutrophil binding fraction (abstract))
  • other P < 0.05 significance threshold (statistical analysis threshold)

Statistical methods review

Model: sonnet

A neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.

The study combined single-cell RNA sequencing of flow-sorted Stx2-B-positive peripheral white blood cells, in vitro cell–toxin binding and co-culture assays, and two murine HUS models (LPS/Stx2 and EDL933 infection). Continuous outcomes from experiments with a minimum of three independent replicates were compared by two-tailed t-test, one-way ANOVA with Tukey post-hoc correction, or two-way ANOVA with Bonferroni post-hoc correction as appropriate; survival data were analyzed by log-rank (Mantel-Cox) test. Results are expressed as mean ± SEM with a significance threshold of P < 0.05.

Replicationbiological Sample sizeMinimum of three independent experiments stated; per-group mouse n not reported in the available text; scRNA-seq performed on cells pooled from one described donor preparation (8681 quality-filtered cells) GroupsLPS/Stx2-treated mice versus CD14 neutralizing antibody or bindarit intervention groups; in vitro cell-line conditions (THP-1, HRGEC, HeLa, PANC-1); human peripheral blood donors (healthy volunteers) Pairingunclear Randomization/blindingnot stated DispersionSEM Effect sizesno Confidence intervalsno Multiplicity correctionTukey HSD (post one-way ANOVA) and Bonferroni (post two-way ANOVA); no correction described for any standalone t-tests
Statistical tests used
Test Applied to n Assumptions
Two-tailed Student's t-test Pairwise comparisons throughout in vitro and in vivo experiments (specific figures not enumerated in the statistical methods section) not stated
One-way ANOVA followed by Tukey's multiple comparisons test Multi-group comparisons among three or more conditions (specific figures not enumerated in the statistical methods section) not stated
Two-way ANOVA followed by Bonferroni's multiple comparisons test Comparisons involving two independent factors (specific figures not enumerated in the statistical methods section) not stated
Log-rank (Mantel-Cox) test Survival curve analysis in mouse HUS models not stated
Approaches that could also have been used
  • Variability is reported exclusively as SEM throughout all experiments
    Could also: Report SD or 95% confidence intervals alongside or instead of SEM — With as few as three independent replicates, SEM is numerically very narrow relative to SD; SD directly describes the spread of the observed data values, while 95% CIs convey both the precision of the mean estimate and a plausible range, making between-group variability and biological heterogeneity more transparent to readers
  • Parametric tests (t-test, ANOVA) were used as the sole inferential framework without reporting normality or variance-homogeneity checks
    Could also: Apply non-parametric alternatives (Mann-Whitney U, Kruskal-Wallis with Dunn's post-hoc) or report Shapiro-Wilk and Levene tests to document that parametric assumptions were evaluated — With n = 3 independent replicates, normality is effectively untestable; non-parametric rank-based methods make no distributional assumptions and are commonly used as a conservative option for small-n biological assay data
  • Multiple standalone t-tests appear to be used alongside ANOVA analyses without description of how the overall family-wise error rate is controlled across all tests in the paper
    Could also: Consolidate all multi-group comparisons under a single ANOVA framework, or apply a global false-discovery rate correction (e.g., Benjamini-Hochberg) across the full set of hypothesis tests — Running many independent t-tests and ANOVA families without accounting for the total number of comparisons can cumulatively inflate the type I error rate; a unified or global correction approach makes the inferential scope explicit
  • scRNA-seq cell-type proportions were described descriptively (e.g., monocytes occupied ~70% of Stx2-B-bound cells) without formal statistical testing of compositional differences
    Could also: Apply compositional statistical analysis (e.g., scCODA, Dirichlet-multinomial regression) to compare cell-type proportions with quantified uncertainty — Standard Seurat clustering characterizes cell-type composition but does not formally test whether proportional differences are statistically reliable; compositional methods account for the sum-to-one constraint inherent in proportion data and produce credible intervals or p-values for abundance shifts
  • Sample size was described as 'a minimum of three independent experiments' without a formal a priori power calculation
    Could also: Conduct and report an a priori power analysis specifying the expected effect size, alpha level, and target power — Pre-specified power calculations make the rationale for the chosen n transparent and allow readers to assess whether the study was adequately powered to detect biologically meaningful differences; this is particularly relevant for in vivo endpoints where animal numbers are ethically constrained
  • The scRNA-seq experiment was conducted on Stx2-B-positive cells sorted from a single donor preparation, and no statistical framework was described for comparing gene-expression or cluster proportions across donors
    Could also: Include cells from multiple independent donors and apply a pseudobulk differential expression approach (e.g., DESeq2 or edgeR on aggregated per-donor counts) to account for donor-level variability — Pseudobulk methods treat the biological replicate (donor) as the unit of analysis rather than individual cells, avoiding the inflated degrees of freedom that arise when thousands of correlated cells are treated as independent observations, and are now widely recommended for single-cell studies with multiple subjects
Software: R/Seurat 3.0 · 10x Cell Ranger 1.2.0 · R/ggplot2 · ImageJ · AlphaFold Server · PyMOL · BioRender

Citation network

Where this publication sits in the reproducibility-weighted citation graph — what it is built on, and what is built on it. Citation data from OpenAlex.

Citations
3
Impact: low
Foundation confidence
None of its references are in our reproducibility record yet — its foundation cannot be assessed.
Topics

No assessed neighbours yet — the network grows as more papers are assessed.

Data lineage

The datasets this paper uses (text-mined from the full text via Europe PMC), and which other assessed papers stand on the same data. A shared dataset is a factual link — not a judgement.

O00206 UniProt in Results (http://purl.org/orb/Results)
also used by 1 paper:
C34554 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
GSE180476 GEO in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
GSE202109 GEO in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
P09386 UniProt in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet

What was reproduced

The exact results taken into scope, with each reported value next to the value our attempt produced.

Scope — PMID 39871175

Title: Monocytes serve as Shiga toxin carriers during the development of hemolytic uremic syndrome. PMCID: PMC11773931 · DOI: 10.1186/s11658-025-00689-8 Code: https://github.com/Xinlei672/Stx2-cell (own code, loose R/Seurat script; commit pushed 2023-12-30) Data (real, scRNA): GEO GSE252352 / GSM8000021 — "Stx2 + cell", human peripheral white blood cells sorted for Stx2-B binding, 10x Genomics 3' v2, made Public on 2026-06-01. Standard CellRanger output: barcodes.tsv.gz, features.tsv.gz, matrix.mtx.gz (~80 MB RAW.tar).

NOTE: the brief listed geo:GSE202109, but that is a cited reference dataset (McEvoy et al. healthy human kidney atlas, PMID 36496458) — not this paper's own data. The paper's Data-availability statement names GSE252352 as the scRNA data; GSE202109 + GSE180476 are external comparison atlases. We reproduce against GSE252352.

In scope (pipeline-derived, single-cell)

The paper's Fig.1 scRNA analysis of Stx2+ sorted white blood cells is a standard Seurat (v3) pipeline, fully specified in the repo stx2_code script and Methods:

  • Load 10x matrix → CreateSeuratObject(min.cells=3, min.features=200)
  • QC: percent.mt < 10 (+ ERCC, nFeature_RNA > 200)
  • Norm/Scale/PCA, FindVariableFeatures(vst, 1500), dims 1:20
  • Doublet removal (DoubletFinder, 7.5% expected rate)
  • Clustering FindClusters (paper: resolution 1.2; repo script: 0.6)
  • Cell-type annotation clusters → Monocyte / Neutrophil / NKT cell / Mast cell via canonical markers (Monocyte: CD14,LYZ,CST3; Neutrophil: CXCR2,S100A8,S100A9; NKT: CD3D,CD3E,CD8A,KLRD1,NKG7,CCL5; Mast: CPA3,CCR3)

Reproducible target numbers (see claims.tsv):

  • C1: ~9410 cell barcodes (pre-filter)
  • C2: 8681 quality-filtered cells (after QC)
  • C3: monocytes ≈ 70% of all Stx2-B-bound cells (dominant population)
  • C4: neutrophils < 30%

Out of scope (not attempted — wet-lab / manual / external)

  • Stx2-B-FITC flow-cytometry binding assays ("nearly all monocytes bind Stx2", "<10% of neutrophils") — wet-lab cytometry, not a pipeline output.
  • Western blots, in-vivo mouse HUS, kidney histology, uncropped blot images.
  • ccl_heatmap script: operates on a local hand-curated ccltotal/ccl chemokine table that is not shipped in repo or GEO → not reproducible (no data).
  • GENELIST.xlsx DotPlot panel: file «path» not shipped; we substitute the canonical marker panel hard-coded later in the same script.

Reproduction strategy

Run the repo's Seurat pipeline on GSM8000021 on «our HPC» (conda Seurat env in-job), reporting pre/post-QC cell counts, cluster count, and cell-type composition. 80/20: exact byte-identical clustering is not expected (Seurat/DoubletFinder stochasticity, version drift v3→present); the robust comparison is cell counts and the monocyte-dominant composition (~70%).

C1
Reported
9410 cell barcodes
Reproduced
9410
exact
C2
Reported
8681 quality-filtered cells
Reproduced
9276 (without DoubletFinder)
partial
C3
Reported
monocytes ~70% (nearly 70%) of Stx2-B-bound cells
Reproduced
70.47%
within tolerance
C4
Reported
neutrophils <30%
Reproduced
27.49%
within tolerance
C5
Reported
cell types: Monocyte, Neutrophil, NKT cell, Mast cell
Reproduced
Monocyte, Neutrophil, NKT_cell, Mast_cell
exact

Assessments & scoring basis

Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.

🤖 AI curator · claude (ai-curator room) · v1.0 L1 84/100

An automated assessment. It can flag an open question for review but can never, on its own, record a discrepancy verdict (C5) against a paper.

🟢1. Data identity
🟢2. Endpoint comparability
🟡3. Location of the main deviation
🟢4. Cause of the deviation
🟢5. Derivability / plausibility
🟢6. Severity of the deviation
🟢7. Core claim
🟢8. Severity of the miss (overall human judgment)
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Concordant (toward reproduced)
All content-critical questions reproduced
-4 pts
From: Q7 · Core claim 🟢
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score -5

This study reproduces essentially 1:1: the central thesis that monocytes are ~70% of Stx2-B-bound cells reproduces to 70.47% (neutrophils 27.49%, <30%), the 9410 pre-filter barcode count is byte-exact, and all four cell lineages match. The only deviation is the post-QC count (9276 vs 8681), which is fully explained on our side by the un-installable DoubletFinder step (9276·0.925≈8580≈8681) — a tooling/version constraint, not an authors' or data defect. No fabrication concern: every value is directly derivable from the shipped GEO data and repo pipeline.

🤝
Reproduced automatically — and fairly

Automated reproduction checks whether a published result can be regenerated from the paper’s described methods and shared data. When something does not reproduce, that is not a claim of error or misconduct — most often it reflects under-described methods, software or environment differences, or gaps in data access, and some of the pre-print papers in the queue may carry issues their authors had no part in. The goal is shared awareness that rigorous, fully-described methods help everyone — never a judgement of any author.

Are you an author? We would genuinely like to hear from you — to clarify the record, add data or code, re-run the pipeline after an accession update, and publish your response right next to the assessment. Everything here is open and auditable.

🚩 Report an error in this record

Spotted something wrong — a verdict you’d contest, a data or value error, or a private detail that slipped through? Tell us, with a short justification. Authors and readers are equally welcome to write in; we review every report.

Prefer email, or the form below not working? Contact us at support@doesitreproduce.com.

Reproduction footprint

claude-opus-4-8

Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.

126.6 k
tokens (I/O) · 8.2 M incl. cache
16 min
runtime · 0.02 CPU-h
6.1 GB
peak RAM
1
HPC jobs
hummel
machine