miR-4478 Accelerates Nucleus Pulposus Cells Apoptosis Induced by Oxidative Stress by Targeting MTH1.
The main results reproduced: recomputed values matched the published ones within tolerance.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
- ✓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
- Every checked point held up.
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 -> clean 1:1. The repo (articalcode/Spine) ships its pipeline as a knitted R notebook (mirna代码.html) with output tables embedded, so reported values are directly readable. The bioinformatic core is a two-dataset miRNA-microarray limma DE (GEOquery getGEO -> auto-log2 -> normalizeBetweenArrays -> lmFit/contrasts/eBayes -> topTable) on GSE63492 (5 control/5 degenerated) and GSE116726 (3 traumatic/3 degeneration), then intersection of P<0.05 & |logFC|>=1 hits. Re-running it on «our HPC» (R 4.3.3, GEOquery 2.70.0, limma 3.58.1) reproduced the 10-miRNA merged-DEG table to all printed digits (logFC, P.Value, t, AveExpr, adj.P.Val, B) and the 6-miRNA consistent-up set including hsa-miR-4478 — exact, no fabrication concern. One trivial code fix: makeContrasts(<string>, levels=) must pass the contrast via contrasts= on current limma (mathematically identical). NOT attempted (out of scope): downstream target prediction (TargetScan7.2/miRWalk/miRPathDB -> 122 targets of miR-486-5p) + GO/KEGG, because the required external DB export files are not shipped in the repo and are unversioned (not byte-reproducible); and all wet-lab (qRT-PCR, luciferase miR-4478->MTH1, apoptosis, IHC), which is non-computational. Note: GSE70362 named in the brief is NOT used by the shipped 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.
-
v1 current initial assessment Score 100assessed: 2026-06-15 ⛓ 99a205edcfca
✎ 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.
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
- 2026-06-15
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-15no 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: opusThe study tests whether differentially expressed miRNAs—specifically miR-4478—contribute to intervertebral disk degeneration (IVDD) by regulating oxidative stress-induced apoptosis of nucleus pulposus cells, with MTH1 as a candidate target gene.
- ★ miR-4478 is upregulated in nucleus pulposus tissues from IVDD patients and is degeneration-degree-associated finding
- ★ Silencing of miR-4478 inhibits H2O2-induced nucleus pulposus cell apoptosis finding
- ★ MTH1 (NUDT1) is a direct target gene of miR-4478 mechanism
- ★ miR-4478 regulates H2O2-induced NPC apoptosis by modulating MTH1-mediated oxidative stress mechanism
- ★ Downregulation of miR-4478 via inhibitor injection alleviates IVDD in a mouse model finding
- miR-4478 has therapeutic potential as a target for oxidative stress-induced IVDD treatment finding
- Bioinformatic integration of GEO microarray datasets with WGCNA and target-prediction tools identified miR-4478 and its targets method
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| miRNA/mRNA microarray bioinformatics analysis (DEM identification, WGCNA, GO/KEGG, target prediction) | Human NP/annulus disk tissue datasets (GSE70362, GSE116726, GSE63492) | none | differentially expressed miRNAs/mRNAs and predicted miR-4478 targets | Affymetrix HG-U133 Plus 2.0; Agilent-070156 Human miRNA; Exiqon miRCURY LNA array; R v4.0.0, TargetScan/miRWalk/miRDB |
| Quantitative real-time PCR (qRT-PCR) | Human NP tissues and cultured human NPCs | none / IVDD vs control | miR-4478 and MTH1 relative expression (2^-ΔΔCt), normalized to U6/GAPDH | TRIzol; Takara RR047A; Sangon B532451-0020 |
| Fluorescence in situ hybridization (FISH) | NP tissues from IVDD patients | none | miR-4478 localization/expression | Exiqon LNA probe; TSA Plus Fluorescein System (PerkinElmer); Olympus IX-81 FV1000 confocal |
| Immunohistochemical staining | Human and mouse disks | none / IVDD model | percentage of MTH1+ cells | anti-MTH1 ab197028 (Abcam); Olympus BX63; ImageJ |
| Flow cytometry (Annexin V-FITC apoptosis detection) | Human NPCs | H2O2 (100 μM, 12 h) ± miR-4478 mimics/inhibitor or MTH1 siRNA | apoptosis rate | Annexin V-FITC Apoptosis Detection Kit (BD) |
| Western blot | Human NPCs | H2O2 induction; miR-4478 mimics/inhibitor; MTH1 siRNA | MTH1 and apoptosis-related protein levels | — |
| Dual-luciferase reporter assay | Human NPCs | co-transfection of MTH1 3'UTR reporter + miR-4478 | luciferase activity to confirm miR-4478 binding to MTH1 3'UTR | Dual-Luciferase Reporter Assay System (Promega) |
| CCK8 viability assay | Human primary NPCs (Procell CP-H097) | H2O2 treatment | cell viability (absorbance at 450 nm) | CCK8 (Dojindo) |
- ▲ miR-4478 upregulated in IVDD NP tissues vs control
- ▲ hsa-miR-4478 differentially expressed in GSE63492 1.24586-fold
- ▲ hsa-miR-4478 differentially expressed in GSE116726 1.715781-fold
- ▼ Silencing miR-4478 reduces H2O2-induced NPC apoptosis
- – MTH1 identified as a target of miR-4478 (predicted by TargetScan and miRDB) CWC score −0.31
- ▼ miR-4478 inhibitor injection alleviates IVDD in mouse model
- fold_change 1.715781 (hsa-miR-4478 fold change in GSE116726 (3 control vs 3 IVDD))
- pvalue 4.76E−05 (hsa-miR-4478 P value in GSE116726)
- fold_change 1.24586 (hsa-miR-4478 fold change in GSE63492 (5 control vs 5 IVDD))
- pvalue 0.011455 (hsa-miR-4478 P value in GSE63492)
- count 24 (human disk samples (12 degenerative herniation, 12 trauma controls))
- count 122 (genes from WGCNA compiled with miR-4478 targets)
- other 100 μM H2O2 for 12 hours (oxidative stress induction condition for NPCs)
- other ≥50% positive staining = high MTH1 expression (IHC MTH1 scoring threshold)
Statistical methods review
Model: sonnetA neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.
The paper combines bioinformatics analysis of public GEO miRNA microarray datasets with in vitro cell culture experiments and an in vivo mouse model to characterize miR-4478 in intervertebral disk degeneration. Differentially expressed miRNAs were identified from two GEO datasets via quantile normalization, fold-change filtering, and volcano plot analysis in R v4.0.0; WGCNA and GO/KEGG enrichment analyses were used to select the candidate target gene MTH1. Clinical validation was conducted in 12 IVDD versus 12 control nucleus pulposus tissue samples; functional validation used H2O2-treated nucleus pulposus cells assessed by flow cytometry, CCK8, western blot, and dual-luciferase reporter assay, and a mouse IVDD model. The specific inferential statistical tests applied to the cell-level and in vivo comparisons are not described in the available methods text.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| Fold change filtering with volcano plot; underlying statistical model not stated (p-values reported in Table 3 without specifying test) | Identification of differentially expressed miRNAs in GSE116726 and GSE63492 miRNA microarray datasets | GSE116726: 3 control vs 3 IVDD; GSE63492: 5 control vs 5 IVDD | not stated |
| Weighted Gene Co-expression Network Analysis (WGCNA) | Module identification and module-trait relationship analysis of mRNA microarray dataset GSE70362 | 14 control vs 10 IVDD | not stated |
| GO and KEGG enrichment analysis via ClusterProfiler and DOSE R packages (hypergeometric test by default; method not explicitly stated) | Pathway analysis of 122 WGCNA-identified genes intersected with predicted miR-4478 target genes | 122 genes | not stated |
| Not stated in available text | Flow cytometry apoptosis assay, CCK8 viability assay, western blot quantification, dual-luciferase reporter assay, and in vivo mouse experiments | — | not stated |
-
Differentially expressed miRNAs were selected via fold change filtering and volcano plots; the statistical model underlying the reported p-values is not described↳ Could also: A moderated t-test framework such as limma's eBayes with Benjamini-Hochberg FDR adjustment could also be applied to the same microarray data — Moderated t-tests stabilize variance estimates by borrowing information across probes, which is particularly relevant when group sizes are very small (n=3 or n=5); explicit FDR control makes the expected false-discovery rate among reported hits transparent
-
Two miRNA microarray datasets (GSE116726 and GSE63492) were analyzed independently and their top dysregulated miRNAs were compared by inspection to select concordant candidates↳ Could also: A formal cross-study meta-analysis (e.g., RankProd, Fisher's combined p-value, or random-effects meta-analysis of log fold changes) could also pool evidence across both datasets — Meta-analytic methods yield a single quantitative evidence measure that accounts for inter-study heterogeneity and the differing sample sizes of the two datasets, providing a more formal basis for prioritizing candidates consistent across studies
-
MTH1 immunohistochemical expression in tissue sections was dichotomized at a 50% positive-cell cutoff to classify samples as low vs high expression↳ Could also: Retaining the percentage of MTH1-positive cells as a continuous variable and comparing groups with a Mann-Whitney U test or correlating with Pfirrmann grade using Spearman's rho could also be used — Continuous analysis preserves information and avoids sensitivity to the choice of cutoff; with n=24 and non-normal ordinal-adjacent data, non-parametric approaches are well suited
-
Pfirrmann grading (I–V) was used to stratify patients into two binary groups (control: I/II vs degeneration: III/IV/V)↳ Could also: Treating Pfirrmann grade as an ordered categorical variable and applying Spearman correlation or ordinal regression of miR-4478 expression against grade could also be used — Preserving the ordinal scale allows examination of a dose-response relationship between miR-4478 expression and degeneration severity rather than collapsing it to a binary contrast
-
Multiple experimental conditions (mimic, inhibitor, siRNA knockdown, rescue combinations) across multiple readouts (apoptosis rate, viability, protein levels, luciferase activity) were assessed, with the statistical approach not stated in the available text↳ Could also: A one-way ANOVA with a Tukey HSD or Dunnett's post-hoc correction could also be applied across the multi-group cell experiments — An omnibus test followed by a multiplicity-corrected post-hoc procedure controls the family-wise error rate across pairwise comparisons within each experimental readout, complementing the overall inference
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.
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.
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — pmid-36130054
Paper: miR-4478 Accelerates Nucleus Pulposus Cells Apoptosis Induced by Oxidative
Stress by Targeting MTH1. PMID 36130054 · PMC9897280 · DOI 10.1097/brs.0000000000004486.
Code: https://github.com/articalcode/Spine @ main (pushed 2022-06-13, GPLv3).
The repo ships ONE artifact: mirna代码.html — a knitted R notebook ("mirna code demo")
that is the bioinformatic pipeline, with its output tables embedded in the HTML
(so the reported intermediate values are directly readable).
What the pipeline actually does (ground truth = the repo HTML, not the brief)
The brief lists data = GSE70362, but the repo code does NOT use GSE70362. The miRNA-screening pipeline uses two miRNA microarray GEO series:
- GSE63492 (GPL19449, 10 samples) — "Revisiting the miRNA expression profiling of
human IDD"; group =
condition:ch1control vs degenerated. - GSE116726 (GPL20712 Agilent miRNA, 6 samples) — "Targeting miR-141 in IDD"; group = traumatic_lumbar_fracture (3) vs intervertebral_disc_degeneration (3). (GSE70362 is the mRNA IVD set from a different paper; not touched by the shipped code.)
Per-dataset pipeline (identical for both): GEOquery::getGEO → (GSE63492 only: probe→
miRNA annotation via GPL miRNA_ID_LIST, collapse by max rowMeans) → auto log2 (GEO2R
quantile heuristic) → limma::normalizeBetweenArrays → lmFit/contrasts.fit/eBayes
(no trend) → topTable(n=Inf). Then joint analysis: filter each by
P.Value<0.05 & |logFC|>=1, merge by miRNA name → 10 common miRNAs (Fig.1 / the
embedded "合并后的差异基因" table). Then keep same-sign-positive trend → 6 "up" miRNAs
(incl. hsa-miR-4478, the paper's focus).
IN SCOPE (pipeline-derived, deterministic, clearly specified — attempt)
- C1 GSE63492 limma logFC + P.Value for the reported miRNAs (the
.xcolumns). - C2 GSE116726 limma logFC + P.Value for the reported miRNAs (the
.ycolumns). - C3 The intersection: the set of 10 common DEG miRNAs (names) from the two filters.
- C4 The 6 consistent-trend "up" miRNAs (same sign, positive), including miR-4478.
Anchor (reported values) = the table embedded in mirna代码.html, transcribed into
original/claims.tsv.
OUT OF SCOPE (the hard/under-specified ~20% — not attempted, stated why)
- Target prediction (TargetScan7.2 / miRWalk / miRPathDB → 122 targets of
hsa-miR-486-5p) and the downstream GO/KEGG enrichment: depend on external static
DB export files (
TargetScan7.2__*.txt,miRWalk_*.csv,miRPathDB_*.csv) that are NOT shipped in the repo and whose download dates/versions are unpinned → not byte-reproducible. Also they target miR-486-5p, not the paper's miR-4478. - Wet-lab (qRT-PCR, luciferase miR-4478→MTH1 binding, apoptosis assays, IHC) — not computational. Out of scope by design.
- GSE70362 — not used by the shipped code; ignored.
Assessments & scoring basis
Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.
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.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
Clean 1:1 reproduction. Re-running the repo's shipped two-dataset miRNA limma DE pipeline on public GEO data (GSE63492, GSE116726) reproduced the 10-miRNA merged-DEG table and the 6-miRNA consistent-up set including miR-4478 to every printed digit; the only code change was a mathematically-identical makeContrasts fix. No discrepancy on any side — deviations are pure display rounding (e.g. 1.24586 vs 1.245860). The downstream target prediction + GO/KEGG and all wet-lab mechanism (luciferase miR-4478→MTH1, apoptosis) were out of scope/non-computational, so the computational claims feeding miR-4478 selection are fully confirmed while the broader biological conclusion is simply untested here, not contradicted.
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-4-8Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.