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

Alginate oligosaccharides improve germ cell development and testicular microenvironment to rescue busulfan disrupted spermatogenesis.

Theranostics · 2020
L1 79/100 3/4
Why this verdict

Part of the results reproduced; minor but material deviations remained.

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)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q4 · Cause of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +7
✓ What held up
  • Same input data as the authors
  • Reported values were directly comparable
What did not (or only partly)
  • 🟡A deviation arose in the data or preprocessing
  • 🟡A deviation was attributed to the published material
  • 🟡Reported values were not (fully) derivable from the shared data
  • 🟡The deviation was non-trivial in magnitude
  • 🟡The central claim did not (fully) hold under reproduction
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
79/100
Reproducibility score
0.3 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 55% of all assessed papers rank 514 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

salvaged by watchdog from agreement.json (agent omitted ROOM_RESULT.json)

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.

✎ 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
2026-06-19
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-18
no human curator yet
Last updated
2026-07-29

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 study tests whether alginate oligosaccharides (AOS) can mitigate busulfan-induced disruption of spermatogenesis and rescue germ cell development and the testicular microenvironment in mice, examined at the single-cell level.

Core claims
  • AOS rescues busulfan-disrupted spermatogenesis in vivo by increasing the proportion of germ cells (spermatocytes and spermatids) finding
  • AOS increases sperm motility and concentration in busulfan-treated mice finding
  • AOS modifies transcription factors/regulons (e.g., Jund, Wt1, Lef1, Elf2) to regulate gene expression and rescue spermatogenesis mechanism
  • AOS promotes ex vivo expression of genes important for spermatogenesis (e.g., PRM1, PRM2, TNP1, TNP2, ODF1) finding
  • AOS improves blood and testicular metabolomes and gut microbiota to support recovery of spermatogenesis finding
  • AOS alters key proteins involved in germ cell development and stress (increased piwil1, ZFP37, p-ERK; decreased Gpx1 and caspase 8) in busulfan-treated testes mechanism
  • Single-cell RNA sequencing can resolve testicular germ and somatic cell populations to assess spermatogenesis rescue method
  • AOS could be used to improve fertility in patients undergoing chemotherapy and combat other causes of infertility finding
Experimental setups
Assay System Perturbation Readout Platform
single-cell RNA-seq (10x scRNA-seq) murine (ICR mouse) testis tubular cells, in vivo busulfan (40 mg/kg) and/or AOS (10 mg/kg BW) cell type proportions, gene expression, regulon activity 10x Genomics
computer-assisted sperm assay (CASA) mouse spermatozoa busulfan and/or AOS sperm motility and concentration CASA
immunofluorescence staining mouse testis samples busulfan and/or AOS marker-positive cell counts (DDX4, DAZL, SYCP3, TNP1, SOX9)
Western blotting mouse testis busulfan and/or AOS protein levels (Jund, Wt1, piwil1, ZFP37, p-ERK, Gpx1, caspase 8)
bulk RNA-seq cultured mouse testis tissue, ex vivo busulfan in vivo then AOS in culture (10 or 50 µg/mL), 48 h differentially expressed genes
LC/MS metabolomics mouse plasma (blood) and testicular tissue homogenate busulfan and/or AOS blood and testis metabolites liquid chromatography-mass spectrometry
16S rRNA sequencing mouse intestinal digesta busulfan and/or AOS gut microbiota composition
histopathological analysis mouse testis busulfan and/or AOS testicular histology
Key results
  • BA10 increased sperm motility versus busulfan alone (B0) 4.2-fold
  • BA10 increased sperm concentration versus busulfan alone (B0) 3.1-fold
  • Spermatocyte (SPC) proportion: A0 58.92%, A10 48.46%, B0 13.87%, BA10 67.87% — AOS restored SPCs after busulfan 13.87% to 67.87%
  • Spermatid (ST) proportion: A0 29.48%, A10 40.67%, B0 0.36%, BA10 12.44% — AOS increased STs after busulfan 0.36% to 12.44%
  • SPG proportion lowest in B0 (8.50%) and highest in BA10 (19.64%) 8.50% to 19.64%
  • A10 modestly increased sperm motility and concentration versus A0 113.8% motility; 116.8% concentration
  • 185 differentially active regulons identified across SPG/SPC/ST clusters; Jund enriched in A10, Wt1 in BA10 185 regulons
  • Ex vivo BA10 showed 345 genes up and 835 down versus B0; up-regulated genes enriched for spermatogenesis 345 up / 835 down
Key statistics
  • fold_change 4.2-fold (sperm motility increase BA10 vs B0)
  • fold_change 3.1-fold (sperm concentration increase BA10 vs B0)
  • count 8941, 9682, 4659, 3778 cells (testicular cells obtained for A0, A10, B0, BA10 scRNA-seq)
  • count 27060 cells in 16 clusters (total cells organized for scRNA-seq clustering)
  • count 185 regulons (differentially active regulons in SCENIC analysis)
  • count 345 up / 835 down (DEGs BA10(Ex) vs B0(Ex))
  • count 681 up / 808 down (DEGs BA50(Ex) vs B0(Ex))
  • other 574 SPG, 629 SPC, 497 ST, 446 LC/SC marker genes (marker genes per cluster for enrichment analysis)

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.

The study used a four-group in vivo design (vehicle, AOS, busulfan, busulfan+AOS) plus a six-group ex vivo culture design, profiling testes by 10x single-cell RNA sequencing, bulk RNA sequencing, and LC/MS metabolomics, with confirmation by histology, immunofluorescence, and Western blotting. High-dimensional data were analyzed with standard pipelines (Seurat clustering and t-SNE, Monocle pseudotime, SCENIC regulon inference, and GO/protein-protein interaction enrichment reporting p-values). Phenotypic outcomes such as sperm motility and concentration were reported as percentages and fold-changes between groups and described as significant, though the specific inferential tests and sample sizes are not detailed in the available text.

Replicationunclear Groups4 in vivo groups (A0/A10/B0/BA10) and 6 ex vivo groups (A0/A10/A50/B0/BA10/BA50 Ex) Pairingunpaired Randomization/blindingnot stated Dispersionunclear Effect sizesyes Confidence intervalsno
Statistical tests used
Test Applied to n Assumptions
unspecified significance test for sperm motility/concentration (described as 'significantly' increased) Figure 1B-C, sperm motility and concentration by CASA across A0/A10/B0/BA10 not stated
differential expression analysis for scRNA-seq cluster marker genes (Seurat) Figure 1E-F, S1, S3; identification of cluster marker genes not stated
SCENIC AUCell regulon activity analysis Figure 2I-J; 185 regulons across SPG/SPC/ST clusters na
bulk RNA-seq differential expression (up/down-regulated gene calls) Figure 4A-B; ex vivo comparisons A0 vs A10/A50, B0 vs BA10/BA50 not stated
GO / pathway and protein-protein interaction (MCODE) enrichment analysis reporting p-values Figure 3A-E, Figure 4C-E; functional enrichment of marker/DE genes na
Approaches that could also have been used
  • Group differences in sperm motility and concentration were described as significant and summarized with percentages and fold-changes.
    Could also: Reporting the per-group n, the specific test used (e.g., one-way ANOVA across the four groups with a post-hoc procedure such as Tukey HSD), and a dispersion measure (SD, IQR, or a 95% CI). — Naming the test, n, and spread alongside the effect size makes the inferential basis transparent and lets readers gauge precision, and a single ANOVA-plus-post-hoc framework handles all four-group comparisons together.
  • Functional enrichment (GO and MCODE) results were colored/ranked by p-value.
    Could also: Reporting adjusted p-values from a multiple-testing procedure such as Benjamini-Hochberg FDR alongside the raw p-values. — FDR adjustment is standard for the large families of terms tested in enrichment analysis and conveys how the term ranking holds up after accounting for the number of comparisons.
  • Differentially expressed genes in the bulk RNA-seq comparisons were reported as counts of up- and down-regulated genes.
    Could also: Stating the modeling framework and thresholds (e.g., DESeq2 or edgeR/limma-voom with the fold-change and adjusted-p cutoffs used). — Specifying the DE method and cutoffs makes the gene lists reproducible and clarifies how significance and effect size were jointly defined.
  • Cluster identities and marker genes were assigned from Seurat clustering and known markers.
    Could also: Reporting the marker-detection test (e.g., Wilcoxon rank-sum as used by Seurat's FindMarkers) and any minimum fold-change/percentage thresholds. — Documenting the marker test and thresholds clarifies how cluster-defining genes were selected and supports replication of the cell-type annotation.
  • Western blot and immunofluorescence comparisons were described qualitatively (e.g., 'increased', 'more abundant').
    Could also: Quantifying band/signal intensity across biological replicates with a summary statistic and a corresponding test. — Adding densitometry/quantification with replicate-level statistics complements the representative images with a numeric measure of the difference.
Software: R / Seurat (scRNA-seq clustering, t-SNE) · Monocle (pseudotime trajectory) · SCENIC / AUCell (gene regulatory network inference)

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
138
Impact: 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.

What was reproduced

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

Scope — pmid-32194870

Paper: Zhao et al. 2020, Theranostics 10(7):3308–3324. "Alginate oligosaccharides improve germ cell development and testicular microenvironment to rescue busulfan disrupted spermatogenesis." DOI 10.7150/thno.43189 · PMID 32194870 · PMCID PMC7053202.

Code link (registry): https://github.com/aertslab/SCENIC — a third-party tool (gene regulatory network inference). Per BRIEF rule P16, applying SCENIC to the paper's own data is an equally valid reproduction. We additionally reproduce the upstream Seurat clustering that SCENIC and the rest of the scRNA-seq analysis depend on, because those steps produce the clearest, most checkable numbers.

Data: GEO GSE131629. Ships a processed expression matrix GSE131629_Testis.AOS3_data.txt.gz (736 MB) covering all four groups, plus raw FASTQ under SRA SRP199191 / BioProject PRJNA544236. Groups: A0 = AOS0 (GSM3791446), A10 = AOS10 (GSM3791447), B0 = B+A0 / busulfan (GSM3791448), BA10 = B+A10 / busulfan+AOS (GSM3791449).

In scope (pipeline-derived, computational)

# Result Pipeline Where in paper
C1 Total cells analysed = 27,060 CellRanger→Seurat QC Results; design
C2 Per-group cell counts: A0 8941, A10 9682, B0 4659, BA10 3778 CellRanger→Seurat Results
C3 Cells organised into 16 clusters Seurat (t-SNE/clustering) Results, Fig 1D-E
C4 4 major cell-type groups w/ cluster membership: SPG=8,9,10; SPC=0,2,3,6,13,14,15; ST=1,4,7; LC/SC=5,11,12 Seurat marker-based annotation Results, Fig 1
C5 Cell-type proportions per group (Fig 1G): e.g. SPC A0 58.92%, A10 48.46%, B0 13.87%, BA10 67.87%; ST A0 29.48%, A10 40.67%, B0 0.36%, BA10 12.44%; SPG B0 8.50%, BA10 19.64% Seurat Fig 1G
C6 Germ cells subclustered into 13 clusters Seurat (germ-cell subset) Results
C7 Widely-expressed marker-gene counts: SPG 574, SPC 629, ST 497, LC/SC 446 Seurat FindMarkers Results
C8 SCENIC: 185 regulons identified / active across groups SCENIC Results, Fig 2
C9 SCENIC group-enriched TFs: A0/A10 = Klf1, Jund, Sox6; BA10 = Lef1, Elf2; B0 = Wt1, Egr4 SCENIC regulon activity Results, Fig 2I

Out of scope (wet-lab / manual / not computational)

  • Western blot validation of Jund / Wt1 (Fig 2J) — wet-lab.
  • Sperm counts, motility, testis histology, IHC/IF, qPCR, body/organ weights, hormone assays, fertility/litter outcomes — wet-lab.
  • Metascape GO enrichment / PPI networks (Fig 3) — web-tool, manual, not pinnable to a runnable artifact with an exact expected number; not attempted unless time permits as a stretch.

Reproduction strategy

  1. Anchor on the shipped processed matrix (...AOS3_data.txt.gz). It is the exact object the authors clustered, so C1/C2 (cell counts) are checkable directly from its dimensions + cell/group labels — the clearest 1:1 points.
  2. Re-cluster with Seurat following the described workflow (Seurat + t-SNE) → reproduce C3 (16 clusters), C4 (cell-type membership), C5 (proportions). Cluster count depends on resolution; graded with tolerance + documented.
  3. Run SCENIC (the linked tool) on the matrix → C8 (regulon count) and C9 (group-enriched TFs). SCENIC is stochastic + threshold-dependent; graded with tolerance and full honesty about non-determinism.

Heavy compute (matrix load, Seurat, SCENIC + cisTarget mouse DBs) runs on «our HPC» SLURM; all data stays on «infra».

Figures / tables: Fig 1DFig 1Fig 1GFig 2Fig 2I
C1_total_cells
Reported
27060
Reproduced
27060
exact
C2_cells_per_group
Reported
A0 8941 / A10 9682 / B0 4659 / BA10 3778
Reproduced
A0 8941 / A10 9682 / B0 4659 / BA10 3778
exact
C3_n_clusters
Reported
16
Reproduced
16 at res 0.25 (scan brackets 16: res0.2->12, 0.4->20)
within tolerance
C4_celltype_clusters
Reported
4 types: SPG/SPC/ST/LC-SC over 16 clusters
Reproduced
all 4 types recovered; SPG={13}, SPC={1,2,5,7,8,10,12,14}, ST={3,4,6,9,11}, LC/SC={0,15}
within tolerance
C5_proportions
Reported
SPC A0 58.92/A10 48.46/B0 13.87/BA10 67.87; ST A0 29.48/A10 40.67/B0 0.36/BA10 12.44; SPG B0 8.50/BA10 19.64
Reproduced
SPC A0 48.16/A10 47.71/B0 17.15/BA10 95.98; ST A0 51.74/A10 52.25/B0 0.06/BA10 1.03; SPG B0 8.61/BA10 1.56
partial
C6_germ_subclusters
Reported
13
Reproduced
13 at res 0.2 (scan 0.2->13, 0.3->17)
within tolerance
C7_marker_counts
Reported
SPG 574 / SPC 629 / ST 497 / LC-SC 446
Reproduced
SPG 1185 / SPC 0 / ST 1030 / LC-SC 1334
partial
C8_scenic_regulons
Reported
185
Reproduced
PENDING
m.public.grade.uncheckable
C9_scenic_tfs
Reported
A0/A10 Klf1,Jund,Sox6; BA10 Lef1,Elf2; B0 Wt1,Egr4
Reproduced
PENDING
m.public.grade.uncheckable

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 79/100

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)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q4 · Cause of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +7

The shipped GSE131629 matrix lets us reproduce the cell-count claims exactly (C1 27060, C2 per-group counts) and the cluster counts at natural resolutions (C3=16, C6=13), and the central biology — busulfan depleting germ cells and AOS rescuing them — reproduces qualitatively. Deviations are on our methodology / authors' underspecification: the paper gives no clustering resolution, no annotation code, and no marker-gene threshold, so the exact Fig 1G proportions (esp. BA10) and the C7 marker counts diverge (including a SPC=0 artifact and a direction flip in BA10 SPG, 1.56% vs 19.64%). No fabrication signal — numbers plausibly derive from the data — but SCENIC (C8 185 regulons, C9 TFs) was still pending at submission, leaving Fig 2 unverified. Overall a solid-but-partial reproduction with explainable, input-side deviations.

🤝
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.

251.8 k
tokens (I/O) · 22.9 M incl. cache
60 min
runtime
Per-job HPC accounting not captured for this run — the runtime shown is the reproduction’s measured wall-clock time.