Corpus 1,273 assessed · 1,174 scored · 643 reproduced ≥75 · 169 flagged ·∅ 74.1/100
← New search

The rates of adult neurogenesis and oligodendrogenesis are linked to cell cycle regulation through p27-dependent gene repression of SOX2.

Cell Mol Life Sci · 2023
L1 59/100 3/4
Why this verdict

The main results reproduced, with only marginal, non-material deviations.

Reproduced on the brainbox compute brainarbeit.com
✓ What held up
  • Nothing in this column.
What did not (or only partly)
  • 🟡Could not use the authors’ exact input data
  • 🟡Reported values were only indirectly comparable
  • 🟡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
59/100
Reproducibility score
0.9 SD below mean
vs. all fields · 1174 studies
🎯 Scores higher than 19% of all assessed papers rank 926 of 1174 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 to attempt; outcome = PARTIAL with one notable non-reproduction. RNA-seq pipeline (the headline molecular axis) reproduced cleanly from the authors' own GEO count matrix (GSE196328, 10 samples): DESeq2 1.50.2 at FDR<0.05, CPM filter -> 14070 genes, both genotype and differentiation contrasts run. The KO POSITIVE CONTROL is clean (Cdkn1b/p27 strongly DOWN in p27KO, padj 2.7e-11), validating sample labels + pipeline. HOWEVER the headline transcriptomic claim C3 -- 'Sox2 de-repressed (UP) in p27KO' -- does NOT reproduce at the bulk-mRNA level (Sox2 log2FC slightly negative, never significant). This is flagged for human review: the paper most likely demonstrates Sox2 de-repression via protein/ChIP/reporter (wet-lab, out of scope) and bulk neurosphere RNA-seq can dilute cell-type-specific effects -- NOT asserted as fabrication. ATAC-seq (C5/C6) is NOT reproducible because GEO GSE196285 deposits only the 10 MACS2 narrowPeak BEDs (no FASTQ/BAM/bigwig): the Bowtie2/MACS2/DiffBind pipeline and the open/closed differential cannot be re-run; an occupancy characterization was done instead (differentiation reshapes accessibility more than genotype). The DHB-Venus ImageJ tool (C12) is public+present (MIT) with example data; its non-bundled plugin deps were resolved (BioVoxxel) and GUI/Windows-path bugs patched, but the interactive macro hangs in pure headless mode (no xvfb) -- and it maps to no deposited paper figure anyway. NOT attempted: LISA TF-enrichment (external Cistrome DB). All compute ran on «our HPC» SLURM compute nodes; data on «infra»; small results + scripts + logs saved under reproduction/.

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 50
    assessed: 2026-06-19 ⛓ 4420c2e74b5b
✎ 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
2026-06-29
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-19
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

The paper tests whether the cyclin-dependent kinase inhibitor p27 (Cdkn1b), beyond its canonical cell-cycle role, non-canonically regulates gene expression programs (via SOX2 and its downstream targets OLIG2/ASCL1) to coordinate cell cycle exit with differentiation in adult subependymal neurogenesis and oligodendrogenesis.

Core claims
  • p27 restricts residual CDK activity after mitogen withdrawal to antagonize cell cycling, but is not essential for cell cycle exit per se finding
  • p27 is required to repress genes involved in the transit from multipotentiality to differentiation, including the progenitor transcription factors Sox2, Olig2 and Ascl1 finding
  • p27 directly associates with regulatory sequences (promoters/enhancers) in the Sox2, Olig2 and Ascl1 genes mechanism
  • p27 repression of Sox2 hierarchically leads to reduced levels of its downstream targets Olig2 and Ascl1 mechanism
  • In vivo, p27 is required for setting proper SOX2 levels so neuroblasts and oligodendroglial progenitor cells timely exit the cell cycle, in a lineage-dependent manner finding
  • p27-deficient NPCs undergo extra cell divisions but eventually fully differentiate, with a bias toward more neurons and oligodendrocytes but not astrocytes finding
  • Integrated ATAC-seq and RNA-seq profiling identifies p27-dependent chromatin accessibility and gene expression changes at the onset of NPC differentiation method
  • A CDK2 activity reporter (DHB-mVenus) shows p27-deficient cells retain higher residual CDK2 activity at mitogen withdrawal, reducible to WT levels by a CDK1/2 inhibitor finding
Experimental setups
Assay System Perturbation Readout Platform
ATAC-seq adult SEZ-derived NSC/NPC neurosphere cultures (WT and p27KO) mitogen withdrawal (differentiation induction) genome-wide promoter chromatin accessibility (Open/Closed)
bulk RNA-seq adult SEZ-derived NSC/NPC neurosphere cultures (WT and p27KO) mitogen withdrawal (differentiation induction) differential gene expression (UP/DOWN)
LISA integrative TF motif/regulatory modeling (using public MNase-seq/ChIP-seq datasets) computational analysis of RNA-seq/ATAC-seq gene lists none transcription factors associated with Closed/differentially regulated gene promoters LISA bioinformatic tool
HOMER motif discovery computational analysis of Closed+DOWN gene promoter sequences none enriched TF binding motifs HOMER
Immunocytochemistry NSC/NPC cultures (WT and p27KO) mitogen withdrawal; p27 cDNA re-expression (p27-Flag) in KO percentage/intensity of p27, Ki67, SOX2, OLIG2, ASCL1, βIII-tubulin, GFAP, O4 positive cells
EdU incorporation assay NPC cultures (WT and p27KO) mitogen withdrawal percentage of EdU+ (S-phase) cells
Live-cell CDK2 activity reporter imaging (DHB-mVenus) NPC cultures transduced with CSII-EF-DHB-mVenus reporter (WT and p27KO) mitogen withdrawal ± CDK1/2 pharmacological inhibitor cytoplasm/nucleus fluorescence ratio (CDK2 activity) CSII-EF-DHB-mVenus reporter, bioimage analysis
Chromatin immunoprecipitation (ChIP) WT NPC cultures proliferation vs onset of differentiation p27 occupancy at Sox2, Sox2 SRR2, Olig2 and Ascl1 regulatory regions (qPCR, IP/NRA ratio) anti-p27 ChIP-qPCR
Key results
  • 85% reduction in the number of genes with differentiation-associated changes in chromatin accessibility or expression in p27KO vs WT cultures 85% reduction
  • 88% of DE genes showing Closed chromatin during WT differentiation lost this change in p27KO cells 88%
  • p27KO cultures show increased EdU incorporation at 2+1 DIV compared to WT, indicating extra cell cycling after mitogen withdrawal 9.1% ± 2.0 (n=5) vs 2.2% ± 0.4 (n=6)
  • p27 binds Sox2 promoter/SRR2 enhancer and Olig2/Ascl1 promoters in proliferating cells, with binding strengthened at onset of differentiation IP/NRA ratio 1.82-3.70 (proliferation), 2-3-fold increase at differentiation
  • SOX2+OLIG2+ASCL1+ triple-positive cell proportion decreases from proliferation to onset of differentiation in WT cultures 37.3% ± 1.7 to 19.5% ± 2.7 (n=9)
  • p27KO cultures show significantly higher proportion of cells with elevated ASCL1, OLIG2 and SOX2 at onset of differentiation, restored to WT levels by p27 cDNA re-expression
  • p27KO cultures fully differentiate by 2+5 DIV but with a bias toward more neurons and oligodendrocytes, not astrocytes, relative to WT
  • p27-deficient cells show higher residual cytosolic DHB-mVenus signal (CDK2 activity) at 2+1 DIV, normalized to WT levels by CDK1/2 inhibitor treatment
Key statistics
  • pvalue p = 0.0015 (EdU+ cells p27KO (9.1±2.0%, n=5) vs WT (2.2±0.4%, n=6) at 2+1 DIV)
  • fold_change 85% reduction (number of genes with expression/accessibility changes in p27KO vs WT during differentiation)
  • fold_change 88% (proportion of Closed+DE gene changes in WT lost in p27KO)
  • pvalue p = 0.008 (p27), p = 0.016 (Ki67) (repeated measures one-way ANOVA for percentage of p27+ and Ki67+ cells during WT differentiation)
  • mean 37.3% ± 1.7 (2 DIV) vs 19.5% ± 2.7 (2+1 DIV) (percentage of SOX2+OLIG2+ASCL1+ triple-positive cells, p = 0.0015)
  • other ChIP ratio p27-IP/NRA control: 1.82-3.70 (proliferation), strengthened 2- to 3-fold at differentiation (p27 binding to Sox2, Olig2, Ascl1 regulatory regions)

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 paper combines cell-culture-based quantitative assays (immunocytochemistry percentages/intensities, EdU incorporation, a CDK2-activity bioimage reporter) analyzed with ANOVA and pairwise comparisons, with genome-wide ATAC-seq/RNA-seq profiling of wild-type versus p27KO cells analyzed via FDR-controlled differential expression/accessibility calls and bioinformatic TF/motif enrichment tools (LISA, HOMER). Results from cell-based assays are reported as mean ± s.e.m. with exact p-values and significance asterisks, with biological replicates shown as individual dots on graphs.

Replicationbiological Sample sizeSample sizes given per comparison (e.g., n=5, n=6, n=9), shown as individual dots on graphs; no formal power analysis described GroupsWild-type vs p27KO cell cultures; proliferating vs differentiating (mitogen-withdrawal) timepoints; treated vs untreated (CDK1/2 inhibitor) conditions Pairingmixed Randomization/blindingnot stated DispersionSEM Exact p-valuesyes Effect sizesno Confidence intervalsno Multiplicity correctionFDR-based control (referred to as "sFDR-controlled" and "FDR-controlled" p-values; specific procedure such as Benjamini-Hochberg not named)
Statistical tests used
Test Applied to n Assumptions
Repeated measures one-way ANOVA Fig. 1e, percentage of p27+ and Ki67+ cells across differentiation timepoints not stated
Unspecified statistical test (comparison of two group means, p=0.0015) EdU-incorporating cell proportions, p27KO vs wild-type at 2+1 DIV n=5 (p27KO) vs n=6 (wild-type) not stated
Unspecified statistical test (comparison of two group means, p=0.0015) Fig. 3d, percentage of SOX2+OLIG2+ASCL1+ triple-positive cells, 2 DIV vs 2+1 DIV n=9 not stated
FDR-controlled differential expression/accessibility analysis (sFDR) Genome-wide RNA-seq and ATAC-seq comparisons, Fig. 1b–c, 2f–h not stated
LISA integrative TF-association analysis Identification of TFs regulating Closed/DE gene sets, Fig. 1d, Fig. 3b not stated
HOMER motif enrichment analysis (FDR<0.05) TF binding motif enrichment in gene promoters, Fig. 1d, Fig. 3c not stated
Approaches that could also have been used
  • Repeated measures across differentiation timepoints (Fig. 1e) were analyzed with a repeated measures one-way ANOVA.
    Could also: A linear mixed-effects model — Mixed-effects models can accommodate unbalanced replicate numbers or missing timepoints while still accounting for within-culture correlation across repeated measurements.
  • Two-group comparisons (e.g., EdU incorporation, triple-positive cell percentages) appear to use an unspecified parametric test for small sample sizes (n=5–9).
    Could also: A nonparametric test such as the Mann-Whitney U test — With small biological replicate numbers, distributional assumptions underlying parametric tests can be harder to verify, and a rank-based test provides an alternative that does not assume normality.
  • Variability in quantitative measurements is summarized using standard error of the mean (s.e.m.).
    Could also: Reporting standard deviation (SD) or a 95% confidence interval — SD directly conveys the spread of the underlying biological replicates, and a 95% CI additionally communicates the precision and plausible range of the estimated effect, which some readers find complementary to a p-value.
  • Genome-wide differential expression and accessibility analyses use an FDR-controlled threshold, described as "sFDR" without naming the specific procedure.
    Could also: Explicitly specifying and citing the FDR procedure (e.g., Benjamini-Hochberg) — Naming the exact multiple-testing correction procedure can aid reproducibility and let readers relate the reported FDR threshold to a specific, well-characterized statistical method.
  • Multiple individual cell-based comparisons are performed across many figure panels within the same experimental series.
    Could also: A family-wise error correction (e.g., Holm-Bonferroni) applied across the full set of related comparisons in a figure — When many hypothesis tests are shown together, a family-wise correction can help manage the cumulative probability of false-positive findings across the whole comparison set.
  • Statistical significance is conveyed primarily via p-values and threshold asterisks.
    Could also: Reporting standardized effect sizes (e.g., Cohen's d) alongside p-values — Effect sizes provide a measure of the magnitude of the difference independent of sample size, which can complement significance testing, particularly for small-n biological replicate studies.
Software: LISA (bioinformatic TF-association tool) · HOMER (motif discovery tool)

What was reproduced

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

Scope — pmid-36627412

Paper: Domingo-Muelas et al. 2023, Cell Mol Life Sci 80:36. "The rates of adult neurogenesis and oligodendrogenesis are linked to cell cycle regulation through p27-dependent gene repression of SOX2." PMID 36627412 · PMCID PMC9832098 · DOI 10.1007/s00018-022-04676-6.

Artifacts available

Artifact What it is Where Status
paucabar/DHB-Venus ImageJ macro DHB-Venus.ijm + CellProfiler .cpproj + shipped example confocal dataset (Olympus .oif). CDK2-activity (nucleus/cytoplasm ratio) image-analysis tool. GitHub, MIT, commit 542371fb79a14e7317cfe1b6860af638bda2cc81 (2019-09-30, master) public, runnable
GSE196328 RNA-seq of WT & p27KO, 10 samples (GSM5870569–578). SuperSeries member. Supplementary = gene-level raw read count matrix GSE196328_RNAseq_p27_raw_reads_GEO.txt.gz. GEO, public obtainable
GSE196285 ATAC-seq of WT & p27KO, 10 samples (GSM5866907–916). Supplementary = GSE196285_RAW.tar. GEO, public obtainable
GSE196329 SuperSeries wrapping the two above. GEO, public obtainable
MOESM1_ESM.xlsx Supplementary Tables S1+S2 = LISA TF-enrichment p-values on Closed gene sets. Springer ESM (saved in original/) downloaded

Methods (verbatim-derived, pipeline tools)

  • RNA-seq: raw reads → STAR align to Ensembl mouse transcriptome → RSEM count raw reads/gene → filter genes <1 CPM in ≥4 samples → coverage via deepTools bamCoverage (RPKM) → DESeq2 DE, FDR 0.05 up/down. ngsplot metagene/TSS heatmaps.
  • ATAC-seq: FastQC/MultiQC → Bowtie2 map to GRCm38/mm10 (mate dovetailing) → Picard MarkDuplicates → SAMtools CollectInsertSizeMetrics → MACS2 peak calling (narrow, no model, FDR 0.05) → deepTools coverage → DiffBind differential peaks (FDR 0.05; open/closed) → ChIPseeker annotation (Gencode M18, ±3 kb TSS, 5 kb flank).
  • Integration: DiffBind + DESeq2 tables combined in R; genes labeled by Closed_UP/Closed_DOWN/Open_UP/Open_DOWN (Fig 1b/c bubbles + scatter).
  • TF enrichment: LISA (Cistrome) on Closed gene lists → Tables S1/S2.
  • CDK2 imaging: custom ImageJ workflow = the DHB-Venus repo (nucleus & cytoplasm binary masks per cell from DAPI + DHB-mVenus channels → nucleus/cytoplasm intensity ratio).

IN SCOPE (pipeline-derived, reproduction targets — priority order)

  1. RNA-seq DESeq2 differential expression from the provided gene-level count matrix (GSE196328). Skips STAR/RSEM (authors shipped the counts). Reproduce: CPM filter (≥1 CPM in ≥4 samples), DESeq2 DE for the differentiation comparison (PRO vs 2+1 DIV) and the genotype comparison (WT vs p27KO); DEG counts at FDR<0.05; direction/significance of key genes Sox2, Olig2, Ascl1, Cdkn1b(p27). Headline molecular claim: p27 represses Sox2→Olig2/Ascl1 (de-repressed in p27KO).
  2. DHB-Venus ImageJ macro on the repo's shipped example_dataset — run the authors' own published code headless (Fiji) and reproduce the nucleus/cytoplasm ratio results table (tool-level reproducibility; the paper's own images are wet-lab and not deposited, so this validates the code, not the paper figure).
  3. ATAC-seq Bowtie2→MACS2→DiffBind from GSE196285 RAW (stretch; heavier: genome index + alignment). Differential (open/closed) peak counts at FDR 0.05.

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

  • All immunostaining/IHC cell counts, flow cytometry / FACS sorting, p27-ChIP-qPCR, Western blots, qRT-PCR, in-vivo OB/SEZ neurogenesis quantifications, EdU/Ki67 scoring (manual counting "using Olympus and ImageJ/Fiji").
  • LISA TF-enrichment exact p-values (Tables S1/S2): depends on Cistrome DB + the authors' exact upstream gene lists; recorded as reported claims but reproduction is low-tractability (external DB), attempted only if primary targets succeed.
  • "deepTools/ngsplot/IGV" visualization tracks (qualitative, no numeric claim).
Figures / tables: Fig 1Fig 1bTableFig 2a
C1
Reported
RNA-seq DE: DESeq2, FDR 0.05
Reproduced
DESeq2 1.50.2 at FDR<0.05; genotype p27KO-vs-WT 686 DEG (516up/170down), DIFF-vs-PRO 10626 DEG
exact
C2
Reported
CPM filter >=1 in >=4 samples
Reproduced
14070 of 48795 genes pass
exact
C3
Reported
Sox2 de-repressed (UP) in p27KO vs WT
Reproduced
Sox2 NOT significantly up at mRNA level (log2FC -0.12..-0.22, padj 0.36-1.0, slightly negative). KO positive control clean: Cdkn1b/p27 DOWN in p27KO padj 2.7e-11
did not match
C4
Reported
Olig2 & Ascl1 reduced downstream of Sox2
Reproduced
Olig2 below CPM threshold (filtered); Ascl1 not significant in genotype contrasts
partial
C5
Reported
ATAC Bowtie2/MACS2(FDR0.05)/DiffBind(FDR0.05)
Reproduced
Not reproducible: only 10 MACS2 narrowPeak BEDs deposited (no FASTQ/BAM/bigwig). Peak QC: 10 samples, 57.5k-91.8k peaks
partial
C6
Reported
majority DA promoters in DE genes CLOSED at differentiation
Reproduced
Exact DiffBind counts not reproducible; occupancy: differentiation reshapes accessibility more than genotype (Jaccard 0.63-0.65 vs 0.76-0.79)
partial
C12
Reported
DHB-Venus custom ImageJ CDK2 N/C-ratio workflow
Reproduced
Code public+present+MIT; deps resolved (BioVoxxel Classify Particles + Binary Reconstruct); GUI dialogs + Windows path patched; Bio-Formats headless crash bypassed via shipped TIFFs; macro then hangs headless (interactive design, no xvfb). No N/C table; no deposited paper image to compare
partial

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 59/100

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)
🤝
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.

110.3 k
tokens (I/O) · 4.9 M incl. cache
28 min
runtime
Per-job HPC accounting not captured for this run — the runtime shown is the reproduction’s measured wall-clock time.