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

Widespread mono- and oligoadenylation direct small noncoding RNA maturation versus degradation fates.

EMBO J · 2025
L1 78/100 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.

Supporting (toward a concern)
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 +3
✓ What held up
  • Same input data as the authors
  • Reported values were directly comparable
  • Reported values are derivable from the shared data
  • The central claim held under reproduction
What did not (or only partly)
  • 🟡A deviation arose in the data or preprocessing
  • 🟡A deviation was attributed to the published material
  • 🟡The deviation was non-trivial in magnitude
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
78/100
Reproducibility score
at the mean
vs. all fields · 1173 studies
🎯 Scores higher than 51% of all assessed papers rank 533 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 to reproduce the DOWNSTREAM pipeline result 1:1. GEO GSE287260_RAW.tar is the authors' own Tailer output (6 *_linked_tail.csv.gz; columns Sequence,Count,EnsID,Gene_Name,End_Position,Tail_Length,Tail_Sequence), not raw fastq. Recomputing the per-gene post-transcriptional A-tailing fraction (mature-end-independent: A-tail read = Tail_Sequence starts with A; fraction = Count-weighted, mean over 3 reps) on a «our HPC» SLURM job (2180205, pure-stdlib Python 3.12.12, 5 s, exit 0) reproduces the HEADLINE Fig 3E numbers essentially EXACTLY: 7SL1=70.6%, 7SL2=71.3% vs paper '~70%'; and confirms Fig 1B (Pol-III >50% A-tailed when nascent: 7SL1 55.7%, 7SL2 62.4%) and the title's mono-vs-oligo dichotomy (Fig 1C: 7SL/snRNA mean A-run ~1.06-1.11 = mono; sca/snoRNAs 1.36, scaRNA22 2.33 = oligo). The two steady-state genes with highest A-tailing are exactly RN7SL2 then RN7SL1, as the paper claims. NOT attempted (the hard ~20%): (1) raw-read reprocessing fastq->custom-dedup->Cutadapt->STAR 2.7.11b 3-pass->Tailer (custom dedup scripts not public; fastq only on SRA); (2) the authors' interactive per-gene mature_end curation (Shiny app), which is why low-anchor/marginal absolute values differ -- U11 came out 23.4% steady (15.0% nascent) vs reported ~15%, and 7SL3 18.2% (not >50%) -- while the unambiguous 7SL1/7SL2 match exactly. No fabrication signal: all graded values are derivable from the deposited data and the central claim is corroborated to within rounding. Overall status 'partial' is conservative -- the headline quantitative claims reproduced exactly; only curation-dependent absolute values for marginal genes diverge.

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 78
    assessed: 2026-06-16 ⛓ 0b62046bf45e
✎ 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
2026-06-16
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-16
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

To investigate the genome-wide dynamics of human small noncoding RNA (sncRNA) 3'-end processing, the authors test whether post-transcriptional adenylation events are linked to sncRNA transcriptional origin and govern the competition between maturation and degradation fates.

Core claims
  • Newly transcribed human sncRNAs undergo widespread post-transcriptional adenylation in two distinct forms: transient oligoadenylation and stable monoadenylation. finding
  • Oligoadenylation is transient, promoted by TENT4A/4B polymerases, and most commonly found on partially processed unstable snoRNAs, correlating with instability rather than maturation. finding
  • Monoadenylation is broadly catalyzed by TENT2, occurs on Pol-III RNAs and a subset of snRNAs, and stably accumulates to the steady state. mechanism
  • TENT2-mediated monoadenylation inhibits 3'-uridine trimming and extension (uridylation/deuridylation dynamics) of Pol-III RNAs. mechanism
  • Monoadenylation of 7SL RNA prevents its accumulation with nuclear La protein and promotes assembly into cytoplasmic signal recognition particles (SRP). mechanism
  • Genome-wide 3'-end sequencing of newly transcribed (EU-labeled) versus steady-state sncRNAs is a method to globally resolve 3'-end processing dynamics. method
  • Mono- and oligoadenylation have divergent impacts on sncRNA maturation versus degradation fates during biogenesis. finding
Experimental setups
Assay System Perturbation Readout Platform
Global RNA 3'-end sequencing of newly transcribed and steady-state sncRNAs (~90-500 nt) HEK 293T-REx cells 5-ethynyluridine metabolic labeling (2 h) for nascent RNA capture 3'-end nucleotide composition, A-/U-tail frequency and length, 3'-end position/trimming
Nascent RNA capture (click-it chemistry / biotin pulldown) HEK 293T-REx cells 5-ethynyluridine labeling enrichment of newly transcribed sncRNAs over steady state click-it chemistry, biotin pulldown
Differential RNA abundance / stability analysis (DESeq2) HEK 293T-REx cells none log2 ratio of steady-state over newly transcribed sncRNA levels as stability proxy DESeq2
Key results
  • Post-transcriptional A-tailing was prevalent among newly transcribed sncRNAs, especially Pol-III RNAs, some with A-tailing of over 50% of the newly transcribed population >50%
  • snoRNAs showed significantly longer A-tails (oligoadenylation) whereas snRNAs and Pol-III RNAs were typically mono(A)-tailed
  • snoRNA post-transcriptional A- and U-tails were overwhelmingly transient (significant shortening at steady state), while Pol-III RNA tails were generally stable
  • Transiently A-tailed snoRNAs had significantly lower steady-state to newly transcribed RNA ratios, indicating instability
  • Transient snoRNA A-tails were found on snoRNAs not fully processed at their 3'-ends
  • A majority of snRNAs and snoRNAs showed significant 3'-end shortening, while most Pol-III RNAs showed little 3'-end trimming
  • No significant difference in 3'-end trimming between transiently A-/U-tailed sncRNAs and the sncRNA population as a whole, arguing against a general tailing–trimming correlation
  • Transient oligoadenylation was observed on snoRNAs of all types, contrary to a prior report restricting it to H/ACA box snoRNAs
Key statistics
  • count newly transcribed sncRNA tail-percentage analysis (snoRNAs in A-/U-tail percentage analysis (Fig 1B))
  • count snRNAs (snRNAs in newly transcribed A-/U-tail percentage analysis (Fig 1B))
  • count Pol3 RNAs (Pol-III RNAs in newly transcribed A-/U-tail percentage analysis (Fig 1B))
  • count transient A-tails (sncRNAs with transient A-tailing in 3'-trimming cumulative analysis (Fig 2C))
  • count transient U-tails (sncRNAs with transient U-tailing (Fig 2C))
  • count All RNAs (total individual sncRNAs analyzed for stability vs tailing (Fig 2D))
  • count transient A-tail snoRNAs (transiently A-tailed snoRNAs analyzed for stability and tail position (Fig 2E,F))
  • pvalue P < 0.05 (threshold used for KS tests and t tests classifying transient tailing and significant trimming)

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.

This study used genome-wide 3'-end sequencing with n=3 biological replicates to compare sncRNA 3'-end modification frequencies in newly transcribed versus steady-state populations from HEK 293TRex cells. Comparisons of A- and U-tailing distributions across RNA class groups (snoRNAs, snRNAs, Pol-III RNAs) were made with two-sample Kolmogorov-Smirnov tests, deviations from zero were assessed with one-sample two-tailed t-tests, and RNA stability was estimated using DESeq2 with Benjamini-Hochberg-adjusted p-values. Results were displayed as box plots (median, IQR, 1.5×IQR whiskers) and p-values were reported as a binary threshold (bold if P<0.05) without exact values.

Replicationbiological Sample sizen=3 biological replicates stated for 3'-end sequencing libraries; number of RNA species per class stated per figure panel GroupssnoRNAs vs. snRNAs vs. Pol-III RNAs; newly transcribed (EU-labeled, 2 h pulse) vs. steady-state sncRNA populations Pairingunpaired Randomization/blindingnot stated DispersionIQR Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionBenjamini-Hochberg FDR applied within DESeq2 for stability analysis; no correction stated for the series of KS tests and one-sample t-tests
Statistical tests used
Test Applied to n Assumptions
Two-sample Kolmogorov-Smirnov test Comparing percentages of A- and U-tailed sncRNA species and mean tail lengths across transcriptional origin groups (snoRNAs, snRNAs, Pol-III RNAs) — Figs. 1B, 1C; comparing snoRNA stability between transiently A-tailed and non-transiently-A-tailed subsets — Fig. 2E Fig. 1B: snoRNAs n=98, snRNAs n=25, Pol-III n=19; Fig. 1C A-tails: snoRNAs n=74, snRNAs n=19, Pol-III n=15; Fig. 1C U-tails: snoRNAs n=18, snRNAs n=14, Pol-III n=13; Fig. 2E: no A-tail change n=75, transient A-tail n=17 not stated
One-sample two-tailed t-test against mu=0 Log2-fold ratios of A- and U-tailed percentages (steady state over newly transcribed) — Fig. 1D; 3'-end trimming shift from zero — Figs. 2A, 2B; mean position of snoRNA A-tails relative to unadenylated counterparts — Fig. 2F Fig. 2A/2B: snoRNAs n=98, snRNAs n=25, Pol-III n=19; Fig. 2F: n=17; Fig. 1D group sizes not explicitly restated not stated
Two-sample two-tailed t-test Per-RNA-species classification of transient A- or U-tailing (P<0.05 threshold for higher fraction tailed in newly transcribed vs. steady state) — used to define subsets plotted in Fig. 2C All RNAs n=142; transient A-tail n=23; transient U-tail n=5 not stated
DESeq2 Wald test with Benjamini-Hochberg FDR adjustment Quantifying sncRNA stability as log2 ratio of steady-state over newly transcribed counts — Fig. 2D n=142 sncRNA species; n=3 biological replicates per condition not stated
Approaches that could also have been used
  • Two-sample Kolmogorov-Smirnov tests were used to compare distributions of A- and U-tailing percentages across RNA class groups
    Could also: Mann-Whitney U (Wilcoxon rank-sum) test — The Mann-Whitney U test is also non-parametric and distribution-free, focuses on rank differences between two groups, and is more widely reported in molecular biology alongside box-plot medians; it also pairs naturally with standardized effect-size measures such as rank-biserial correlation, which the KS test does not directly provide
  • One-sample two-tailed t-tests against mu=0 were used to assess whether log2-fold ratios and 3'-trimming shift values differed from zero
    Could also: Wilcoxon signed-rank test against a hypothesized median of zero — The Wilcoxon signed-rank test requires no normality assumption for the RNA-level values and is equally standard for testing a location parameter against a null; it is a common alternative when per-group n is small (here groups range from n=5 to n=98 RNA species) and the distributional shape of log2-ratios is uncertain
  • Per-RNA-species two-sample t-tests (P<0.05 threshold applied individually) were used to classify each of ~142 species as having transient tailing
    Could also: A single mixed-effects model or generalized linear model with RNA species as a unit of analysis, followed by one FDR correction over all per-species tests — Running one uncorrected test per species inflates the family-wise error rate across ~142 simultaneous classifications; a BH or Storey q-value correction applied to all per-species p-values is a standard approach to controlling false discovery in multi-RNA genomic screens and would be consistent with the DESeq2 framework used elsewhere in the paper
  • DESeq2 was used to estimate RNA stability from count-based sequencing data comparing two conditions
    Could also: edgeR (negative binomial GLM) or limma-voom (precision-weighted linear models) — Both are widely accepted alternatives for count-based differential abundance in RNA-seq; using a second tool alongside DESeq2 is a common sensitivity check in genomics, as results consistent across methods strengthen confidence in conclusions
  • Results across groups were displayed as box plots showing the median and IQR, without overlaid individual data points
    Could also: Superimpose individual data points (strip/dot plots) on the box plots, or replace with beeswarm plots — With n=3 biological replicates underlying each library, the number of experimental units is small; showing individual points alongside the distributional summary allows readers to directly assess spread and identify potential outliers, which is particularly informative in small-n sequencing studies
  • P-values were reported as a binary threshold (P<0.05 shown in bold) without exact numerical values
    Could also: Report exact p-values (e.g., P=0.018) together with a standardized effect size (e.g., rank-biserial r, Cohen's d, or log2-fold change with CI) — Exact p-values allow readers to gauge the degree of evidence and are required for meta-analyses and replication efforts; pairing them with effect sizes separates statistical significance from biological magnitude, which is especially informative when group sizes vary substantially across the RNA species analyzed
Software: DESeq2 · BioRender

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
0
Impact: low
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.

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.

A16035 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
A63880 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
C10365 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
R71007 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet

What was reproduced

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

Scope — pmid-41350938

Paper: Ocheltree et al. 2025, EMBO J, "Widespread mono- and oligoadenylation direct small noncoding RNA maturation versus degradation fates." PMID 41350938 · PMCID PMC12811392 · DOI 10.1038/s44318-025-00655-2. Code: https://github.com/TimNicholsonShaw/tailer (commit 2b8cc44) + downstream https://github.com/TimNicholsonShaw/tailer-analysis (commit 9cbc932). Data: GEO GSE287260 ([global_RNAseq] sub-series).

Pipeline (from Methods)

fastq → custom-Python 3'-adapter+PCR-dup removal → Cutadapt → STAR 2.7.11b 3-pass alignment to hg38 (5'-hard-clip 10 nt, local re-align to single-gene sequences) → Tailer (global mode) → per-read 3'-end + non-templated tail calls → downstream tailer-analysis (R/Shiny) → per-gene A-tailing figures. sncRNA set = GENCODE V33 snRNA/misc_RNA/rRNA/rRNA_pseudogene/snoRNA/scaRNA/ ribozyme/TERC (tRNA excluded); reads truncated >10 nt from annotated 3' end removed.

In scope (attempted)

The deposited GSE287260_RAW.tar is the Tailer output itself — 6 CSVs (*_linked_tail.csv.gz, 3 nascent + 3 totalRNA replicates) with columns Sequence,Count,EnsID,Gene_Name,End_Position,Tail_Length,Tail_Sequence. This lets us reproduce the downstream, pipeline-derived per-gene A-tailing quantities deterministically from the authors' own processed data:

  • Fig 3E — steady-state (totalRNA) post-transcriptional A-tailing % per gene (named anchors: 7SL1, 7SL2 ≈70%; U11 ≈15%).
  • Fig 1B — newly-transcribed (nascent) A-tailing % (>50% for Pol-III RNAs).
  • Fig 1C — mono(A) (snRNA/Pol-III) vs longer oligo(A) (sno/scaRNA): A-run length.
  • Fig 3C/D — which genes accumulate A-tails on >half their population.

Metric used (mature-end-independent, maps to authors' cumulativeTailPlotter/tail_logo_grapher in tailer-analysis_functions.r): A-tail read = Tail_Sequence begins with "A"; mono-A = Tail_Sequence=="A"; A-tail fraction = Σcount(A-tail) / Σcount(all reads for gene), per replicate, mean±sd over 3 replicates of a condition.

Out of scope (NOT attempted — the hard ~20%)

  • Raw read processing (fastq → STAR 3-pass → Tailer). The custom dedup Python scripts and the exact 3-pass STAR small-RNA-genome construction are not in the public repo; raw fastq are only on SRA. We start from the deposited Tailer CSVs instead (P16: the deposited processed output is authoritative).
  • Per-gene mature_end correction. The Shiny app exposes an interactive, manually-curated per-gene mature-3'-end offset. We did not replicate this curation; our metric is mature-end-independent. This is why low-anchor / marginal absolute values (U11, 7SL3) differ from the figure while the dominant 7SL1/7SL2 values match essentially exactly.
  • All wet-lab results (Northern blots, TENT2 knockouts, EU-labeling biochemistry).
Figures / tables: Fig 3EFig 1BFig 1CFig 3C
fig3e_7sl1
Reported
~70%
Reproduced
70.6% (sd 3.4, n=3 totalRNA reps)
exact
fig3e_7sl2
Reported
~70%
Reproduced
71.3% (sd 4.4, n=3 totalRNA reps)
exact
fig3e_u11
Reported
~15%
Reproduced
23.4% steady-state / 15.0% nascent
partial
fig1b_polIII_nascent
Reported
>50%
Reproduced
7SL1 55.7%, 7SL2 62.4% (>50%)
within tolerance
fig1c_sno_vs_sn
Reported
snoRNA longer A-tails; snRNA/PolIII mono(A)
Reproduced
mean A-run PolIII 1.08 / snRNA 1.11 / sno-scaRNA 1.36; scaRNA22 2.33, scaRNA20 2.53
within tolerance
fig3cd_only_7sl_half
Reported
7SL1/7SL2/7SL3 >50%
Reproduced
only 7SL1 70.6% & 7SL2 71.3% >50%; 7SL3 18.2%
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 78/100

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: 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 +3

Starting from the authors' own deposited Tailer output, the headline Fig 3E numbers (7SL1 70.6%, 7SL2 71.3% vs paper '≈70%') and the title's mono-vs-oligoadenylation dichotomy reproduce essentially exactly, and every graded value is derivable from the shared data — no fabrication signal. The deviations that exist (U11 23.4% vs ~15%; Fig 3C/D 7SL3 18.2% vs >50%) are on our methodology side: we did not replicate the authors' interactive per-gene mature_end curation (Shiny app, not deposited) and used a mature-end-independent metric, which only moves low-anchor/marginal genes. Severity is moderate — the central conclusion holds fully while a secondary claim (7SL3 >50%) does not under our simplified metric. Overall a solid reproduction with explainable, curation-driven 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.

180.7 k
tokens (I/O) · 10.8 M incl. cache
16 min
runtime · 0 CPU-h
0 GB
peak RAM
1
HPC jobs
hummel
machine