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

A choreography of centrosomal mRNAs reveals a conserved localization mechanism involving active polysome transport.

Nat Commun · 2021
L1 96/100 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
96/100
Reproducibility score
1.2 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 91% of all assessed papers rank 92 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 -> faithful 1:1 reproduction of the deterministic, pipeline-derived statistics. Zenodo 4322750 deposits the authors' Big-FISH/Cellpose pipeline plus the cleaned end-of-pipeline per-cell feature table (notebook/features.csv, 54263 cells). Re-running the deposited notebook logic (pandas + scipy Welch t-test) on «our HPC» reproduces EXACTLY: N=54,263 cells; the 8 centrosomal interest genes; and the central Fig-3 result that puromycin (polysome dissociation) significantly reduces the centrosomal mRNA proportion for all 8 centrosomal mRNAs (p down to ~0) but not for controls, while cycloheximide has minimal effect. NOT attempted (out of scope / data not deposited): the upstream image-analysis steps (raw 4-channel smFISH images, Cellpose segmentation, Big-FISH spot detection) producing features.csv -- raw images are not in the deposit; and the 602/728-gene high-throughput screen and Drosophila ortholog experiments (separate data not deposited). This is a P16 reproduction of the deposited features through the deposited pipeline; grades provisional, human must confirm.

💻 Code ↗ 🗄 Data: 10.5281/zenodo.4322750

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 96
    assessed: 2026-06-18 ⛓ 2557cc90a8db
✎ 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
2026-06-18
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: opus
Founding hypothesis

Do human mRNAs encoding centrosomal proteins localize to centrosomes, and if so, what is the mechanism—specifically, is their localization dependent on translation and active transport of polysomes?

Core claims
  • A total of eight human mRNAs (PCNT, NIN, BICD2, CCDC88C, CEP350, HMMR, ASPM, NUMA1) localize at centrosomes. finding
  • Centrosomal mRNAs localize at different cell-cycle stages with a finely regulated, choreographed translational program. finding
  • Localization of all eight centrosomal mRNAs requires translation and the nascent protein (puromycin-sensitive, cycloheximide-insensitive). mechanism
  • ASPM and NUMA1 mRNAs localize via active movement/transport of endogenous polysomes toward the centrosome at mitotic onset. mechanism
  • ASPM polysomes associate with microtubules and localize by motor-driven transport or microtubule pulling. mechanism
  • Drosophila orthologs of the human centrosomal mRNAs also localize to centrosomes and also require translation, indicating a conserved family. finding
  • A high-throughput smFISH (HT-smFISH) method was developed to screen centrosomal protein-coding genes. method
  • ASPM, NUMA1, and HMMR mRNAs localize to the centrosome at the same time as their encoded proteins. finding
Experimental setups
Assay System Perturbation Readout Platform
High-throughput smFISH (HT-smFISH) screen HeLa cells stably expressing Centrin1–GFP none centrosomal mRNA enrichment / localization across interphase and mitosis high-throughput spinning-disk confocal microscope (200 nm lateral, 600 nm axial; oligonucleotide pool of 92,000)
Low-throughput smiFISH (validation) HeLa cells stably expressing Centrin1–GFP none confirmation of centrosomal mRNA localization (Cy3 probes) wide-field microscopy
smFISH against GFP RNA HeLa Kyoto BAC cell lines (ASPM-GFP, NUMA1-GFP, HMMR-GFP) none mRNA localization at mitotic phases correlated with protein localization 44 Cy3-labeled oligonucleotide probes
Two-color smFISH HeLa BAC-GFP cell lines none pairwise co-localization of ASPM, NUMA1, HMMR mRNAs (Cy3 BAC-GFP mRNA + Cy5 endogenous mRNA)
Time-lapse live-cell microscopy HeLa Kyoto BAC cell lines (ASPM-GFP, NUMA1-GFP, HMMR-GFP) none protein expression and localization across the cell cycle
smFISH with translation-inhibitor drug treatment HeLa cells expressing Centrin1–GFP drug (puromycin or cycloheximide, 5–20 min) fraction of cells with localized centrosomal mRNAs automated spinning-disk microscope, 63× oil objective
Automated high-content smFISH quantification HeLa–Centrin1–GFP cells in 96-well plates puromycin / cycloheximide fraction of cells with localized mRNA, mitotic-phase classification by DAPI automated spinning disk microscope, 3D imaging, automated DAPI classifier
smFISH in Drosophila (ortholog localization) Drosophila (orthologs Girdin/Bsg25D/BICD/Plp/Asp/Mud/Cen/Cp110) translation inhibition centrosomal mRNA localization and translation dependence
Key results
  • Eight human mRNAs localize at centrosomes; six newly confirmed (NIN, BICD2, CCDC88C, CEP350, plus PCNT and NUMA1) added to previously known ASPM, HMMR. 8 mRNAs
  • CCDC88C localizes in interphase but not mitosis; PCNT, NIN, BICD2, HMMR, CEP350 localize in interphase and early mitosis then delocalize; NUMA1 and ASPM localize only during mitosis.
  • Puromycin delocalized all eight centrosomal mRNAs while cycloheximide had no effect.
  • A 5-min puromycin incubation was sufficient to inhibit centrosomal localization of ASPM, NUMA1, and HMMR mRNAs at all mitotic phases where they normally localize.
  • ASPM-GFP mRNA and protein enriched together on mitotic centrosomes across all phases of division; NUMA1 and HMMR mRNAs accumulated only in prophase/prometaphase.
  • HMMR centrosomal mRNA/protein re-established at telophase cytokinetic bridges, unlike NUMA1.
  • The three mRNAs (ASPM, NUMA1, HMMR) never perfectly co-localized on centrosomes at any cell-cycle stage, occupying distinct peri-centrosomal regions.
  • Drosophila orthologs of the human centrosomal mRNAs localized to centrosomes in a translation-dependent manner.
Key statistics
  • count 602 genes (genes encoding centrosomal/mitotic spindle proteins screened by HT-smFISH)
  • count 8 mRNAs (human mRNAs localizing at centrosomes)
  • count >40,000 cells (automated analysis of cells for quantitative validation)
  • count 92,000 oligonucleotides (complex oligonucleotide pool used to generate probes)
  • count 50–100 probes (individual probes designed against each mRNA)
  • pvalue <0.0001 (****) (two-sided Fisher's exact test for centrosomal localization significance)
  • count 44 Cy3 probes (oligonucleotide probe set against GFP RNA sequence)
  • other 200 nm lateral, 600 nm axial (3D imaging resolution in the screen)

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 is a high-throughput smFISH screen of 602 centrosomal-protein-coding genes in HeLa cells, followed by lower-throughput validation experiments and drug-treatment assays. The primary inferential method is the two-sided Fisher's exact test applied to proportions of cells showing centrosomal mRNA localization across cell-cycle phases. Results are presented as bar graphs of per-phase localization frequencies with 95% Wilson/Brown binomial confidence intervals and asterisk-coded significance tiers; no explicit multiple-testing correction is described for the series of pairwise comparisons performed.

Replicationbiological Sample sizeThree independent experiments pooled; total cell count per condition shown within bar graphs; automated screen analyzed >40,000 cells GroupsCell-cycle phase categories (interphase sub-stages, prophase, prometaphase, metaphase, anaphase, telophase) and drug-treatment conditions (DMSO vs. puromycin vs. cycloheximide) Pairingunpaired Randomization/blindingnot stated DispersionCI Exact p-valuesno Effect sizesno Confidence intervalsyes Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
Two-sided Fisher's exact test Pairwise comparisons of the proportion of cells showing centrosomal mRNA localization across cell-cycle phases (Figs. 1c, 2b, 2d, 2f and related supplementary figures) Total cell counts per bar from three independent experiments (exact per-bar N indicated in bar graphs but not reported numerically in the provided text); automated analysis used >40,000 cells not stated
Approaches that could also have been used
  • Multiple two-sided Fisher's exact tests were applied across the family of cell-cycle-phase comparisons within each mRNA, and across multiple mRNAs, without a stated multiplicity correction
    Could also: Apply a false-discovery-rate procedure (e.g., Benjamini–Hochberg) or a family-wise correction (e.g., Bonferroni) across the set of pairwise tests within each figure panel — With many simultaneous comparisons, a multiplicity correction would make the claimed significance thresholds directly interpretable at the family level; readers could then distinguish which contrasts are robust after accounting for the number of tests
  • P-values are reported only as asterisk tiers (<0.0001, ~0.0001, <0.01, <0.05) rather than as exact values
    Could also: Report exact p-values (e.g., p = 0.0032) alongside or instead of symbol tiers — Exact p-values allow readers and future meta-analysts to apply their own significance thresholds, perform power calculations, and combine results across studies more precisely
  • The magnitude of the difference in localization frequency between cell-cycle phases is not quantified beyond the p-value
    Could also: Report an odds ratio (or relative risk) with its 95% CI for each pairwise comparison — An effect-size estimate would convey the practical magnitude of the enrichment at centrosomes, which is complementary information to the test of whether the difference is non-zero
  • Proportions of localizing cells from three independent biological experiments were pooled and compared with Fisher's exact test
    Could also: Model experiment-level variation with a mixed-effects logistic regression or a generalized linear mixed model treating experiment as a random effect — When counts are pooled across independent biological replicates, between-experiment variability is not captured; a mixed-effects approach would propagate that variance into the inference, which can matter when replicate-level proportions differ
  • For the drug-treatment comparisons (puromycin vs. cycloheximide vs. control), the analysis appears to involve separate Fisher's exact tests per condition
    Could also: Use a single chi-square test of independence (or log-linear model) across the three treatment groups simultaneously, followed by post-hoc pairwise contrasts — A single omnibus test controls the type-I error for the three-way comparison before drilling into pairs; chi-square and Fisher's exact tests are asymptotically equivalent at large n and the omnibus approach is the conventional framing for multi-group categorical data
  • Centrosomal localization was scored as a binary outcome (localized / not localized) per cell
    Could also: Quantify a continuous enrichment metric (e.g., mRNA spot density within a defined radius of the centrosome marker relative to cytoplasm) and compare groups with a non-parametric rank test (e.g., Mann–Whitney U) or a permutation test — A continuous enrichment score captures gradations in localization strength that a binary call may miss, and can be more sensitive for detecting partial or transient localization differences
Software: Not stated in provided text

What was reproduced

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

Figures / tables: Fig 1Fig 3bFig 3
C1_n_cells
Reported
54,263 segmented cells in total
Reproduced
54263 rows in features.csv
exact
C2_interest_genes
Reported
8 centrosomal mRNAs: NIN, BICD2, CCDC88C, CEP350, PCNT, HMMR, NUMA1, ASPM
Reproduced
interest set = {aspm,bicd2,ccdc88c,cep350,hmmr,nin,numa1,pcnt}
exact
C3_puromycin_abolishes
Reported
Puromycin nearly abolished centrosomal accumulation (Fig 3b/3d); centrosomal mRNAs significantly reduced (****, p<0.0001), controls not
Reproduced
all 8 interest genes: mean proportion_rna_centrosome drops under puro, Welch p ranging 6.7e-14 (aspm) to ~0 (bicd2/nin); controls trim59 p=0.071(ns), ttbk2 p=0.011 — significance pattern matches paper stars
exact
C4_chx_minimal
Reported
Cycloheximide had minimal effects (Fig 3)
Reproduced
most interest genes ns under CHX (nin p=0.34, numa1 p=0.71, pcnt p=0.32, cep350 p=0.18); bicd2/aspm not present for chx-untr or small; no abolition of centrosomal localization
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 96/100

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 faithful 1:1 reproduction of the deterministic, deposited end-of-pipeline statistics: N=54,263 cells matches exactly, the 8 centrosomal genes match 8/8, and the Fig-3 puromycin result (significant centrosomal-localization loss for all 8 interest genes, ns controls; CHX minimal) reproduces exactly from features.csv via the shipped scipy Welch-test logic. All values are derivable from the shared data, with no fabrication signal. The only limitation is scope, not correctness — raw images, the 602/728-gene screen, and Drosophila ortholog experiments were not deposited and were not attempted (a q1/q2 availability matter, not an authors' defect on the reproduced 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.

89.9 k
tokens (I/O) · 4.7 M incl. cache
8 min
runtime · 0 CPU-h
0.3 GB
peak RAM
1
HPC jobs
hummel
machine