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

Inhibition of SARS-CoV-2 Infections in Engineered Human Tissues Using Clinical-Grade Soluble Human ACE2.

Cell · 2020
L1 87/100 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.

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
Every question reproduced
-1 pts
From: “every question reproduced”
Total score -7
✓ What held up
  • Same input data as the authors
  • Reported values were directly comparable
  • No relevant deviation in data/preprocessing
  • 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)
  • Every checked point held up.
How its reproducibility compares
87/100
Reproducibility score
0.7 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 72% of all assessed papers rank 301 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

REPRODUCED (1:1, high fidelity) -- re-run after requeue. The paper (Monteil et al, Cell 2020) is overwhelmingly wet-lab; its single pipeline-derived result is the single-cell RNA-seq analysis of the human kidney organoid (Suppl Fig 2), coded in jpromeror/SC_KidneyOrganoid_ACE2 (commit 591e2d0, author JP Romero). Code and data (GEO GSE147863 -> GSM4447249 filtered .h5, sha256 a44af6d7..., the exact script input) are public. We re-ran the Seurat workflow from the raw h5 on «our HPC» (SLURM «job», n094, 7m13s, ExitCode 0; Seurat 4.3.0.1/sctransform 0.4.2, vst.flavor=v1) and compared cell-by-cell against the authors' shipped final object KidneyOrganoid_MetaData.rds. RESULTS: QC quantile thresholds reproduce EXACTLY (nFeature 488/5653; nCount 667/23108.65 vs paper's truncated 667/23108). The post-QC cell set is BYTE-IDENTICAL: 20673 cells, barcode overlap 20673/20673 with the reference, zero mismatches. Cluster count at res 0.4 is EXACT (15). De-novo clustering agrees with the authors' labels at ARI 0.8476; de-novo markers (8103) recover the expected kidney-organoid cell types (podocytes NPHS2/PODXL, endothelium CLDN5, tubular CLDN4/WFDC2, proliferating TOP2A, renin+ REN, stroma COL1A1/SFRP). ACE2 is detected in a restricted 0.66% of cells, consistent with the paper's motivation. This 2026-06-22 re-run is byte-identical to the prior 2026-06-16 run («job»), confirming determinism. No fabrication indicators: every checked value is regenerable from shipped data+code. The only divergences from the 2020 object (per-cluster sizes, ARI<1) are explained by Seurat/sctransform version drift (2020 = Seurat 3.1.x / sctransform 0.2.x). NOT ATTEMPTED (out of scope): all wet-lab virology/organoid-infection/hrsACE2 experiments. Honest verdict: the in-scope computational pipeline reproduces essentially exactly. Note: a fresh conda env build hit a «infra» disk-quota limit, so a byte-identical shared Seurat 4.3.0.1 env was reused; «our HPC» was reachable throughout.

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 87
    assessed: 2026-06-16 ⛓ 938d5350a9ad
✎ 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
2026-06-22
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-16
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: sonnet
Founding hypothesis

Since ACE2 is the receptor SARS-CoV-2 uses to enter cells, clinical-grade human recombinant soluble ACE2 (hrsACE2) should competitively block SARS-CoV-2 infection of cells and human organoids.

Core claims
  • Clinical-grade hrsACE2 significantly inhibits SARS-CoV-2 infection of Vero-E6 cells in a dose-dependent manner finding
  • Murine recombinant soluble ACE2 (mrsACE2) does not inhibit SARS-CoV-2 infection, showing species specificity of hrsACE2's effect finding
  • SARS-CoV-2 can directly infect engineered human blood vessel (capillary) organoids derived from iPSCs, producing infectious progeny virus finding
  • SARS-CoV-2 can directly infect engineered human kidney organoids derived from embryonic stem cells, producing infectious progeny virus finding
  • hrsACE2 significantly reduces SARS-CoV-2 infection of human blood vessel organoids finding
  • hrsACE2 significantly reduces SARS-CoV-2 infection of human kidney organoids in a dose-dependent manner finding
  • A SARS-CoV-2 isolate (clade A3) was obtained from a Swedish COVID-19 patient and sequenced (GenBank MT093571) resource
  • Single-cell RNA-seq of kidney organoids shows ACE2 expression in proximal tubule and podocyte cell clusters, mirroring native kidney tissue finding
Experimental setups
Assay System Perturbation Readout Platform
Viral isolation, electron microscopy, next-generation sequencing, phylogenetic analysis nasopharyngeal sample from Swedish COVID-19 patient; Vero E6 cells for culture none viral particle morphology, clade assignment
qRT-PCR viral RNA quantification (dose-response) Vero-E6 cells hrsACE2 or mrsACE2 pre-mixed with SARS-CoV-2 (10^3, 10^5, 10^6 PFU) prior to infection intracellular viral RNA at 15 hpi
qRT-PCR viral RNA quantification (progeny virus, inoculum not removed) Vero-E6 cells hrsACE2 or mrsACE2 co-incubated with SARS-CoV-2 for 15 h viral RNA at 15 hpi
Organoid generation and immunostaining (CD31, PDGFRβ, DAPI), light microscopy human iPSC-derived blood vessel (capillary) organoids none vascular structure/morphology
qRT-PCR viral RNA quantification +/- hrsACE2, and re-infection assay of Vero E6 with organoid supernatant human iPSC-derived blood vessel organoids SARS-CoV-2 infection (10^6 particles) with or without hrsACE2 viral RNA at 3/6 dpi; progeny virus infectivity on Vero E6
Organoid generation, confocal microscopy (LTL, nephrin, laminin, DAPI), single-cell RNA-seq human embryonic stem cell-derived kidney organoids none tubular/podocyte marker expression, ACE2 expression by cell cluster
qRT-PCR viral RNA quantification +/- hrsACE2, and re-infection assay of Vero E6 with organoid supernatant human ESC-derived kidney organoids SARS-CoV-2 infection (10^6 particles) with or without hrsACE2 (dose-dependent) viral RNA at 6 dpi; progeny virus infectivity on Vero E6
Key results
  • hrsACE2 reduced SARS-CoV-2 recovery from Vero cells 1,000-5,000-fold
  • mrsACE2 had no significant effect on SARS-CoV-2 infection of Vero-E6 cells
  • hrsACE2 inhibition of Vero-E6 infection was dose-dependent and virus-inoculum-dependent (p<0.01 to p<0.001)
  • Viral RNA in blood vessel organoids increased from day 3 to day 6 post-infection, indicating active replication
  • hrsACE2 significantly decreased SARS-CoV-2 infection of blood vessel organoids (p<0.01)
  • SARS-CoV-2 replicated in kidney organoids, detected by qRT-PCR at 6 dpi
  • hrsACE2 significantly reduced SARS-CoV-2 infection of kidney organoids in a dose-dependent manner (p<0.05)
  • Supernatants from infected blood vessel and kidney organoids efficiently infected naive Vero E6 cells, confirming production of infectious progeny virus
Key statistics
  • fold_change 1,000-5,000-fold reduction in viral recovery (hrsACE2 effect on SARS-CoV-2 recovery from Vero cells)
  • pvalue p < 0.01; p < 0.001 (hrsACE2 dose-dependent inhibition of Vero-E6 infection (Figure 2A, Student's t test))
  • pvalue p < 0.05; p < 0.01 (hrsACE2 effect on progeny virus production in Vero-E6 cells (Figure 2C, Student's t test))
  • pvalue p < 0.01 (hrsACE2 effect on blood vessel organoid infection (Figure 3D, Student's t test))
  • pvalue p < 0.05 (hrsACE2 effect on kidney organoid infection (Figure 4D, Student's t test))
  • other 10^3 PFU (MOI 0.02), 10^5 PFU (MOI 2), 10^6 PFU (MOI 20) (viral inoculum doses used to infect Vero-E6 cells)
  • count 6 days post-infection (time point for kidney organoid viral RNA and progeny virus assessment)
  • other clade A3 (phylogenetic classification of the Swedish SARS-CoV-2 isolate)

Statistical methods review

Model: opus

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

This is an experimental in vitro virology study using cell lines (Vero E6) and human iPSC/ESC-derived blood vessel and kidney organoids to test whether clinical-grade human recombinant soluble ACE2 (hrsACE2) inhibits SARS-CoV-2 infection, with viral RNA quantified by qRT-PCR as the primary readout. Group comparisons (e.g., treated versus untreated at various viral doses/MOIs) were made using Student's t test, with significance reported as thresholded p-value strata (∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001). Results were displayed as bar graphs with mean ± SD.

Replicationunclear GroupshrsACE2-treated vs untreated (and mrsACE2 control) across viral doses/MOIs; cells and organoids Pairingunclear Randomization/blindingnot stated DispersionSD Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
Student's t test Figure 2A, hrsACE2 vs control inhibition of SARS-CoV-2 in Vero-E6 cells across MOIs (qRT-PCR viral RNA) not stated
Student's t test Figure 2C, effect of hrsACE2 on progeny virus in Vero E6 cells (qRT-PCR) not stated
Student's t test Figure 3D, effect of hrsACE2 on SARS-CoV-2 infection of blood vessel organoids (qRT-PCR) not stated
Student's t test Figure 4D, effect of hrsACE2 on SARS-CoV-2 infection of kidney organoids (qRT-PCR) not stated
Approaches that could also have been used
  • Group comparisons were made with Student's t test, and the underlying distribution/normality assumptions were not stated.
    Could also: A nonparametric test such as Mann-Whitney U, or reporting whether a normality/variance check (or Welch's correction) was applied, could also be used. — For small-sample in vitro data where normality is hard to establish, nonparametric or Welch-based approaches can also be appropriate and make the distributional assumptions explicit.
  • Several pairwise t tests were performed across multiple viral doses, conditions, and figures.
    Could also: A single ANOVA (e.g., for the dose/MOI series) followed by a post-hoc correction such as Tukey HSD, or a multiplicity adjustment like Benjamini-Hochberg, could also be applied across the family of comparisons. — An omnibus-plus-post-hoc or FDR approach also controls the family-wise/false-discovery rate when many related comparisons are made together.
  • Dispersion was summarized as mean ± SD with significance shown as p-value thresholds (asterisks).
    Could also: Reporting exact p-values together with a 95% confidence interval or an effect-size estimate could also accompany the means. — Exact p-values, CIs, and effect sizes also convey the magnitude and precision of the effect beyond a binary significance threshold, which is informative for small-n data.
  • The number of replicates (biological versus technical) and the n underlying each comparison were not specified in the available text.
    Could also: Explicitly stating n, the replication structure, and how independent experiments were defined could also be included. — Clarifying replication and n also helps readers interpret the variability shown and the basis of each test.
  • Randomization and blinding were not described.
    Could also: Stating whether allocation/assessment was randomized or blinded (or noting it as not applicable) could also be reported. — Documenting randomization/blinding status, even to note it was not used, also supports transparency about how bias was managed.
  • The dose-dependent response was described qualitatively and tested with pairwise comparisons.
    Could also: A dose-response or regression/trend analysis across hrsACE2 concentrations could also be used to model the relationship. — A trend or dose-response model also quantifies the concentration-effect relationship directly, complementing the pairwise comparisons.

Result convergence & founder nodes

Findings this paper shares with others that ran a comparable experiment. A node’s strength is how many independent papers report it (replication breadth) — not how often it is cited, so a heavily-replicated but under-cited founder still stands out.

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
2,261
Impact: very high
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.

AB_777895 RRID in Article (http://semanticscience.org/resource/SIO_001029)
no other assessed paper uses this yet
AB209299 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
GSE147863 GEO in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
GSM4447249 GEO in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
HPA038360 HPA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
MT093571 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
RRID:AB_2154851 RRID in Article (http://semanticscience.org/resource/SIO_001029)
no other assessed paper uses this yet
RRID:AB_2336559 RRID in Article (http://semanticscience.org/resource/SIO_001029)
no other assessed paper uses this yet
RRID:AB_2675975 RRID in Article (http://semanticscience.org/resource/SIO_001029)
no other assessed paper uses this yet
RRID:AB_477163 RRID in Article (http://semanticscience.org/resource/SIO_001029)
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-32333836

Paper: Monteil V, Kwon H, Prado P, ... Penninger JM. Inhibition of SARS-CoV-2 Infections in Engineered Human Tissues Using Clinical-Grade Soluble Human ACE2. Cell 2020. DOI 10.1016/j.cell.2020.04.004.

Code: https://github.com/jpromeror/SC_KidneyOrganoid_ACE2 (commit 591e2d022a48d151b83ae89bc60a6335ec774d30, 2020-04-29). Author of the single-cell analysis code: Juan P. Romero («email»).

Data: GEO GSE147863 → GSE147863_RAW.tar (96 MB) contains exactly one file, GSM4447249_KidneyOrganoid_FilteredGeneBCMatrices.h5 (Cell Ranger filtered gene–barcode matrix). This is the exact input the script reads (script hard-codes a local path .../KidneyOrganoid_FilteredGeneBCMatrices.h5).

What the paper is, and what is computational

This is overwhelmingly a wet-lab paper: SARS-CoV-2 infection of Vero cells, engineered human blood-vessel and kidney organoids, and inhibition by clinical-grade soluble human ACE2 (hrsACE2). The vast majority of reported results (viral-load qPCR, plaque assays, organoid infection, hrsACE2 dose response) are experimental and OUT OF SCOPE — not derivable from any bioinformatic pipeline.

The single, self-contained computational pipeline in this paper is the single-cell RNA-seq analysis of the human kidney organoid (Supplementary Figure 2 — the repo ships it as SuppFigure2.png). It establishes which kidney organoid cell types express ACE2 (the SARS-CoV-2 receptor), motivating the kidney-organoid infection experiments.

IN SCOPE (pipeline-derived — attempted)

The Seurat (v3-era) workflow in KidneyOrganoidAnalysis.R:

  1. Read10X_h5CreateSeuratObject(min.cells=3, min.features=400)
  2. QC: percent.mt = ^MT-; thresholds from 2.5/97.5% quantiles of nCount/nFeature.
  3. Filter: nFeature_RNA 488–5653 & nCount_RNA 667–23108 & percent.mt < 50.
  4. NormalizeDataScaleDataCellCycleScoringSCTransform(vars.to.regress = S.Score, G2M.Score, percent.mt, nFeature_RNA).
  5. RunPCAFindNeighbors(dims=1:20)FindClusters(resolution=0.4)RunUMAP(dims=1:20).
  6. FindAllMarkers; feature plots for ACE2 and lineage markers (SLC3A1, SLC27A2, PODXL, NPHS2, NPHS1, CLDN4, MAL, CD93).

Reference / ground truth shipped in the repo: KidneyOrganoid_MetaData.rds — a data.frame with the per-cell metadata of the final Seurat object plus the UMAP coordinates. This lets us compare a from-scratch re-run against the authors' own object cell-by-cell.

OUT OF SCOPE (wet-lab / not pipeline — not attempted)

All virology and organoid-infection experiments; hrsACE2 production and dose response; qPCR viral RNA quantification; plaque assays; immunostaining. These are the paper's headline results but contain no reproducible computational pipeline.

Reproducibility strategy

  • Deterministic, version-independent claims first: QC quantile thresholds and cell count after filtering depend only on the raw counts + the stated parameters, not on the Seurat version.
  • Version-sensitive claims: SCTransform/clustering/UMAP depend on Seurat & sctransform versions (2020 = Seurat 3.1.x / sctransform 0.2.x). We report the cluster count and an ARI of our clustering vs. the shipped reference labels on shared cells, and flag version sensitivity explicitly.
Figures / tables: Figure2Fig2
C1
Reported
488 / 5653
Reproduced
488 / 5653
exact
C2
Reported
667 / 23108
Reproduced
667 / 23108.65
within tolerance
C3
Reported
20673 cells
Reproduced
20673 cells (barcode overlap 20673/20673, 0 mismatch)
exact
C4
Reported
15 clusters (res 0.4)
Reproduced
15 clusters
exact
C5
Reported
ACE2 restricted/subset expression
Reproduced
137/20673 = 0.66% ACE2+
partial
C6
Reported
n/a (clustering-agreement metric)
Reproduced
ARI 0.8476 vs authors' labels (20673 shared cells)
within tolerance

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 87/100

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.

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
Every question reproduced
-1 pts
From: “every question reproduced”
Total score -7

This is a near-perfect 1:1 reproduction of the only computational result in an otherwise wet-lab Cell paper. The scRNA-seq pipeline reruns from public GEO data (GSE147863) and public code to a byte-identical post-QC cell set (20673/20673 barcodes, 0 mismatch), exact QC thresholds, exact 15-cluster count, and ARI 0.848 vs the authors' shipped labels. The only deviations (C2's truncated 23108 vs 23108.65, per-cluster sizes, ARI<1) sit on the technical/expected side — Seurat/sctransform version drift — not the authors' or our methodology. Caveat: the paper's headline hrsACE2-inhibition conclusions are wet-lab and out of scope, so q7 reflects only the confirmed computational claims.

🤝
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.

277.9 k
tokens (I/O) · 27.2 M incl. cache
113 min
runtime · 0.19 CPU-h
25.3 GB
peak RAM
1
HPC jobs
hummel
machine