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

Interaction between SNAI2 and MYOD enhances oncogenesis and suppresses differentiation in Fusion Negative Rhabdomyosarcoma.

Nat Commun · 2021
50/100 3/4
Why this verdict

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

Reproduced on the brainbox compute brainarbeit.com
How its reproducibility compares
50/100
Reproducibility score
1.4 SD below mean
vs. all fields · 1173 studies
🎯 Scores higher than 8% of all assessed papers rank 1026 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

PARTIAL (honest), clean re-run after requeue. Full pipeline re-run end-to-end on the paper's own data on «our HPC» («job», node n096): GSE137168 SMS-CTR H3K27ac SRR10092020 (38,990,976 raw reads, matches SRA) + matched input SRR10092024 (25,223,241 raw, matches) -> BWA hg19 -> samtools markdup -r -> MACS2 2.1.4 p<=1e-10 -> younglab/ROSE -g HG19 -t 0 (12.5kb stitch). Run with TWO aligner readings of the paper's unspecified 'BWA': bwa aln+samse (era-appropriate single-end, primary) gave 34,388 peaks -> 640 super-enhancers / 13,521 typical; bwa mem gave 38,666 peaks -> 665 super / 14,845 typical. Reported (Fig. S6c) = 516 super / 8,223 typical. Both aligners reproduce the qualitative + order-of-magnitude result (a few hundred SEs, thousands of typicals) and grade PARTIAL on both claims; bwa aln is modestly closer (SE ratio 1.24 vs 1.29, typical 1.64 vs 1.81), showing the aligner is a minor not dominant driver. The bwa mem result reproduces a prior INDEPENDENT run (archived 2026-06-20, different job) BIT-FOR-BIT (665/14845/38666/15510) -> determinism confirmed. No evidence of fabrication: the reported 516/8223 are plausible outputs of this exact pipeline under a more conservative peak set / the bamliquidator backend / blacklist filtering. Two pipeline bugs found and fixed along the way (a streaming single-end dedup pipe that dropped the dedup BAM; a cross-sample BAM mixup that made MACS2 see treatment==control -> 0 peaks), both now guarded (sequential dedup + a treat!=control read-count assertion) and recorded in the kartei. NOT attempted (out of scope): wet-lab assays, SNAI2/MYOD ChIP overlap/motif, RNA-seq DE, HiChIP loops.

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: 2026-06-20 ⛓ 775bf9153533
✎ 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
2026-06-23
Rubric version
not recorded
Assessed by
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 SNAI2, highly expressed in Fusion Negative Rhabdomyosarcoma (FN-RMS), interacts with the myogenic master regulator MYOD to block terminal muscle differentiation and thereby promote oncogenesis.

Core claims
  • SNAI2 is highly expressed in FN-RMS tumors and cell lines compared to normal tissue finding
  • SNAI2 is oncogenic, blocks myogenic differentiation, and promotes growth in FN-RMS finding
  • MYOD directly activates SNAI2 transcription via super enhancers with a striped 3D contact architecture mechanism
  • SNAI2 preferentially binds enhancer elements and competes with MYOD at a subset of myogenic enhancers required for terminal differentiation mechanism
  • SNAI2 suppresses a muscle differentiation program modulated by MYOG, MEF2, and CDKN1A mechanism
  • RAS/MEK signaling modulates SNAI2 levels and chromatin binding, linking oncogenic RAS-driven differentiation blockade to SNAI2 mechanism
  • Combining SNAI2 reduction with vincristine synergistically reduces tumor volume and enhances muscle differentiation finding
  • SNAI2 knockdown reduces tumorigenicity, proliferation, and stemness while inducing differentiation in vitro and in vivo finding
Experimental setups
Assay System Perturbation Readout Platform
RNA-seq expression analysis RMS tumor samples and cell lines vs normal tissue none SNAI2 mRNA expression (log2 FPKM)
Western blot RMS cell lines vs primary human myoblasts none SNAI2 protein level
Immunohistochemistry FN-RMS and FP-RMS primary pediatric tumors none SNAI2 protein staining positivity
ChIP-seq (H3K27ac and MYOD) FP-RMS/FN-RMS cell lines and primary tumors, myoblasts, myotubes, skeletal muscle none super enhancer and MYOD binding at SNAI2 locus
Hi-C / HiChIP chromatin conformation IMR90 fibroblasts (Hi-C); SMS-CTR FN-RMS cells (HiChIP H3K27ac) none 3D topological/enhancer interaction structure at SNAI2 TAD
siRNA knockdown with western blot and qRT-PCR FN-RMS cell lines (RD, JR1, SMS-CTR) MYOD1 siRNA knockdown SNAI2 mRNA and protein expression
CRISPRi (dCas9-KRAB) with qRT-PCR SMS-CTR FN-RMS cells sgRNA-targeted suppression of MYOD-bound enhancers E1-E5 near SNAI2 SNAI2 mRNA expression
shRNA knockdown with immunostaining, qRT-PCR, western blot, proliferation/colony/sphere assays, and xenograft tumor growth FN-RMS cell lines (RD, JR1, SMS-CTR, RD18) in vitro and RD xenografts in mice, ± vincristine shRNA SNAI2 knockdown, combined with vincristine in vivo differentiation markers (MyHC, MEF2C), proliferation/colony/sphere counts, tumor volume and weight
Key results
  • SNAI2 expression is elevated in RMS tumor samples/cell lines vs normal tissue, highest in FN-RMS >4 log2 FPKM
  • MYOD1 and SNAI2 mRNA expression positively correlate in RMS and normal muscle but not other tissues Pearson r=0.428 (FN-RMS)
  • siRNA knockdown of MYOD1 suppresses SNAI2 protein and mRNA in RD, JR1, and SMS-CTR cells
  • CRISPRi silencing of MYOD-bound enhancers E1 and E2 (highest MYOD signal) strongly reduces SNAI2 transcription; distal E3-E5 sites do not
  • SNAI2 knockdown in RD cells in differentiation medium increases MyHC-positive cells ~20-fold (shSNAI2.1), ~8-fold (shSNAI2.2)
  • SNAI2 knockdown reduces rhabdosphere formation in RD, JR1, and SMS-CTR cells e.g. RD: 345.6±19.7 to 167.6±3.5 spheres
  • SNAI2 knockdown reduces colony forming units and soft agar colonies in RD cells 3-fold reduction in CFUs
  • SNAI2 knockdown combined with vincristine synergistically reduces xenograft tumor volume and enhances differentiation vs either alone
Key statistics
  • correlation Pearson 0.428, p=0.0000038 (MYOD1 vs SNAI2 expression correlation in FN-RMS)
  • correlation Pearson 0.372, p=0.0001 (MYOD1 vs SNAI2 expression correlation in FP-RMS)
  • correlation Pearson 0.494, p=0.001 (MYOD1 vs SNAI2 expression correlation in normal muscle)
  • correlation Pearson 0.051, p=0.477 (MYOD1 vs SNAI2 expression correlation in other (non-muscle) tissues)
  • count 14 of 19 positive (SNAI2 IHC positivity in FN-RMS primary tumors)
  • count 3 of 4 positive (SNAI2 IHC positivity in FP-RMS primary tumors)
  • fold_change ~20-fold (shSNAI2.1) and ~8-fold (shSNAI2.2) (Increase in MyHC-positive RD cells after SNAI2 knockdown in differentiation medium)
  • mean shScr 1.19%±0.41 vs shSNAI2.1 20.75%±2.39 MyHC+ cells (Quantitation of MyHC-positive RD cells after SNAI2 knockdown)

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 primarily employed Student's two-tailed t-tests to compare shRNA knockdown versus scramble control groups across a wide range of in vitro and in vivo assays including qRT-PCR, immunostaining quantitation, proliferation curves, colony formation, sphere formation, and xenograft tumor volume and weight. A Welch's-corrected t-test was used for one CRISPRi perturbation comparison. Pearson correlations were computed for MYOD1/SNAI2 co-expression across RNA-seq tumor and tissue cohorts. Results were consistently reported as mean ± SD with exact p-values; no multiplicity correction was described for the many parallel comparisons performed.

Replicationbiological Sample sizeStated as 'n=3 biologically independent experiments' for most in vitro assays; per-arm group sizes specified for xenograft experiments (n=3–6 per arm in initial experiment; n=10–20 per arm in vincristine experiment); no formal a priori power calculation described GroupsshSNAI2 (two independent hairpins: shSNAI2.1 and shSNAI2.2) vs. scramble shRNA controls, across multiple FN-RMS cell lines (RD, JR1, SMS-CTR, RD18); in vivo xenograft arms with and without vincristine co-treatment Pairingunpaired Randomization/blindingnot stated DispersionSD Exact p-valuesyes Effect sizesno Confidence intervalsno Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
Student's two-tailed t-test Multiple comparisons throughout: qRT-PCR gene expression (Figs. 1h, 2d, 2e, 2i, 3a–e, 3g, 3h, 3o, 3p), MyHC+ immunostaining quantitation, proliferation growth curves, colony forming units, soft agar colony counts, sphere formation counts, xenograft tumor volume and tumor weight n=3 biologically independent experiments for most in vitro assays; n=6 shScr, n=3 shSNAI2.1, n=3 shSNAI2.2 for initial xenograft experiment; n=10 shSNAI2, n=20 shScr for vincristine xenograft not stated
Student's t-test with Welch's correction CRISPRi dCas9-KRAB guide RNA perturbation of SNAI2 enhancer regions E1–E5 in SMS-CTR cells (Fig. 1i) n=3 biologically independent replicates not stated
Pearson correlation Co-expression analysis of MYOD1 and SNAI2 across FN-RMS, FP-RMS, muscle, and non-muscle tissue RNA-seq cohorts (Fig. S1c) Cohort sizes not explicitly stated; implied by magnitude of reported p-values not stated
Approaches that could also have been used
  • Multiple pairwise Student's t-tests were conducted across many assay types, cell lines, and conditions without a multiplicity correction procedure
    Could also: A one-way ANOVA followed by a post-hoc correction (e.g., Dunnett's test against a common scramble control, or Tukey HSD) could also be used when three or more groups are compared within a single experiment — ANOVA with post-hoc correction provides one standard way to account for the inflation of type I error when multiple group means are tested simultaneously, and Dunnett's test is specifically designed for the many-vs-one-control structure common in knockdown experiments
  • Dispersion is reported as SD throughout, including for experiments with n=3 independent biological replicates
    Could also: A 95% confidence interval could also be reported alongside or instead of SD — Confidence intervals directly convey the uncertainty around the estimated group mean and are interpretable in terms of population-level inference; at small n, CI width makes the precision of the estimate visually explicit in a way that SD alone does not
  • Student's equal-variance t-test was used as the primary two-group test throughout; Welch's correction was applied in one instance (Fig. 1i)
    Could also: Welch's t-test (unequal-variance form) could be applied consistently as the default two-group test — Welch's t-test does not assume equal group variances and performs similarly to Student's t-test when variances happen to be equal, making it a robust default especially when group sizes or variances may differ across experiments
  • Pearson correlation was used to assess MYOD1/SNAI2 co-expression in RNA-seq cohorts
    Could also: Spearman rank correlation could also be used for the same purpose — Spearman correlation makes no assumption of bivariate normality and is less sensitive to outliers or extreme values, both of which are common features of RNA-seq-derived expression distributions
  • Immunohistochemistry results in primary tumors were reported descriptively as counts of positive cases (e.g., 14 of 19 FN-RMS, 3 of 4 FP-RMS) without a formal inferential test
    Could also: A Fisher's exact test could also be applied to compare proportions of SNAI2-positive tumors between FN-RMS and FP-RMS subtypes — Fisher's exact test is well-suited to small contingency tables and would provide a p-value for the difference in positivity rates between subtypes, complementing the descriptive counts
  • Longitudinal xenograft tumor volume data were compared between arms using t-tests applied at individual time points
    Could also: A linear mixed-effects model or repeated-measures ANOVA could also be applied to model the full tumor growth trajectory — These approaches use all longitudinal measurements jointly, account for within-animal correlation across time points, and can estimate the overall treatment effect on growth rate rather than requiring separate tests at each time point
Software: Not explicitly stated for statistical analyses in the available text; RNA-seq expression values referenced as FPKM from previously published datasets

What was reproduced

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

Scope — pmid-33420019

Paper: Pomella et al. 2021, Nat Commun 12:192. "Interaction between SNAI2 and MYOD enhances oncogenesis and suppresses differentiation in Fusion Negative Rhabdomyosarcoma." DOI 10.1038/s41467-020-20386-8. Code: https://github.com/linlabbcm/rose2 (ROSE2, v1.1.0, Oct 2019) — a third-party super-enhancer caller (P16: equally valid to reproduce by applying the named tool to the paper's data). Data: GEO GSE137168 (SRA SRP221199).

Pipeline-derived results (in scope)

The paper's only explicit, pinnable ROSE2-derived numbers:

"Using the ROSE2 (Rank Order of Super Enhancers) algorithm, we classified the enhancers into typical (n = 8223) and super (n = 516, SEs)" — Fig. S6c, SMS-CTR cells, from H3K27ac ChIP-seq.

Reproduction target (RU claim C1/C2):

  • C1: number of typical enhancers in SMS-CTR = 8223
  • C2: number of super-enhancers in SMS-CTR = 516

Pipeline as described in Methods

  • Aligner: BWA, genome build hg19.
  • Peak calling: MACS2.1, H3K27ac peaks at stringent p ≤ 1e-10.
  • Enhancer calling: ROSE2 (parameters not stated → tool defaults: stitching distance 12.5 kb, TSS exclusion -t 0, genome HG19).
  • Super-enhancer = points above the inflection of the ranked H3K27ac signal curve.

Input samples (GSE137168)

  • H3K27ac (rankby): GSM4072342 "CTR shCtrl H3K27ac" = SRX6824671 = SRR10092020 (single-end, 39.0M reads)
  • Input (control): GSM4072346 "CTR shCtrl input" = SRX6824675 = SRR10092024 (single-end, 25.2M reads)

"CTR" in the GEO naming = SMS-CTR cell line; shCtrl = baseline (non-targeting shRNA).

Out of scope (not attempted)

  • Wet-lab assays (qPCR, westerns, IHC, xenografts, proliferation, differentiation) — manual.
  • SNAI2/MYOD ChIP-seq peak overlap analyses, motif analyses, GSEA, RNA-seq DE — separate pipelines; the headline ROSE2 SE/TE count is the cleanest 80% target.
  • HiChIP loop calling (GSM4948219) — different pipeline, not the ROSE2 claim.
  • The hard last ~20% (exact peak-set reproduction depends on undisclosed MACS2 flags beyond p-value; ROSE2 params not stated) is acknowledged, not chased.

Reproduction strategy

Single SLURM job on «our HPC» std: conda env (sra-tools, bwa, samtools, macs2, rose2+bamliquidator) → fetch FASTQ from ENA → BWA aln hg19 → MACS2 (p 1e-10) → GFF → ROSE2 -g HG19 -t 0 → count SE vs TE → compare to 516 / 8223. All data on «infra»; only counts + logs returned to «host».

Figures / tables: Fig. S6c
C1
Reported
typical enhancers in SMS-CTR = 8223 (Fig. S6c)
Reproduced
13521 (bwa aln, primary; ratio 1.64) / 14845 (bwa mem; ratio 1.81) — same ballpark (thousands), runs high
partial
C2
Reported
super-enhancers in SMS-CTR = 516 (Fig. S6c)
Reproduced
640 (bwa aln, primary; ratio 1.24) / 665 (bwa mem; ratio 1.29) — a few hundred SEs in both; independently re-counted from the SE table
partial

Assessments & scoring basis

Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.

No assessment has been recorded yet.
🤝
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

<synthetic>

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.

1.2 M
tokens (I/O) · 89.1 M incl. cache
547 min
runtime · 18.02 CPU-h
29.6 GB
peak RAM
3 (1 failed)
HPC jobs
hummel
machine