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

ZAP targets aberrant mRNA transcripts encoding proteins with defective signal peptides for degradation.

EMBO J · 2026
L1 50/100 3/4
Why this verdict

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

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: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
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 🟡
Minor / cosmetic deviation
+1 pts
From: Q2 · Endpoint comparability 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +8
✓ 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
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

Described well enough to attempt. Primary clean target = RNA-seq differential expression: paper specifies NO RNA-seq aligner/quantifier/DE params, so a standard third-party pipeline (salmon 1.10.3 + GENCODE v44 + tximport + DESeq2 1.50.2) was applied to the paper's own PRJNA1265689 fastqs (P16-valid). Contrast PPLmutZapKO vs PPLmut (ZAP-KO vs ZAP-intact, matched defective-signal-peptide reporter) = deposited proxy for Fig 5A 'ZAP KO vs parental K562'. «our HPC» «job» ran end-to-end (COMPLETED 32:48): all 6 RNA-seq samples downloaded+quantified (mapping 40-74%), DESeq2 gave 1150 up / 913 down genes. C1: ATF5 reproduced UP and highly significant (5.88-fold, log2FC 2.56, padj 1.2e-15) but ~3.4x below the reported ~20-fold -> partial (lower magnitude expected for a reporter-line proxy contrast + different quantifier). C2: 35.5% of up genes endomembrane/secretory vs reported ~26% (same order of magnitude, 1.19x enriched over background) -> partial (GO approximation, not authors' exact gene set). DIFFERENT-but-consistent reproduction, not 1:1. NOT attempted: C3 eCLIP 1233 crosslink sites (optional hard-20%), Fig 4C 7SL enrichment, ribosome profiling, MassIVE proteomics, reporter wet-lab. Nothing fabricated; all grades provisional for human sign-off.

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: 2026-06-16 ⛓ 52431ffbc536
✎ 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
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: sonnet
Founding hypothesis

The molecular components and mechanism of the regulation of aberrant protein production (RAPP) pathway, which detects nascent proteins with defective signal peptides and degrades their mRNA templates, are largely unknown; this study tests whether ZAP is a core RAPP effector that couples SRP-mediated signal peptide recognition to aberrant mRNA degradation.

Core claims
  • ZAP (ZC3HAV1/PARP13) is a key component of the RAPP quality control pathway finding
  • The short isoform ZAP-S associates with SRP complex components and facilitates degradation of aberrant mRNAs encoding defective signal peptides mechanism
  • Loss of ZAP activates the unfolded protein response (UPR) and the downstream integrated stress response (ISR) finding
  • AGO2 is required for RAPP activity on the mutant PPL reporter, validating prior reports finding
  • RAPP reduces both mRNA stability and translation efficiency (ribosome loading) of aberrant signal-peptide-containing transcripts finding
  • Canonical RIDD and ribosome quality control factors (ERN1/IRE1, PELO, HBS1L, N4BP2) are not required for RAPP finding
  • An inducible dual-fluorescence PPL-mCherry/EGFP reporter system was developed to study RAPP resource
  • STK11 was identified as a screen hit but showed little functional effect on RAPP reporter stabilization upon validation finding
Experimental setups
Assay System Perturbation Readout Platform
RT-qPCR K562 and HeLa cells with PPL-WT/∆2L/∆4L reporters signal peptide leucine deletion (∆2L, ∆4L) mCherry mRNA levels normalized to EGFP
Western blot / immunoblotting K562 and HeLa cells with PPL reporters; cells and culture media signal peptide leucine deletion; sgRNA knockdown (ZAP, SRP54) mCherry and EGFP protein levels, secreted vs intracellular species
Immunofluorescence microscopy K562 cells with PPL-WT/∆2L reporter signal peptide leucine deletion colocalization of reporter with calreticulin (ER marker), Pearson's r
Flow cytometry K562 stable cell lines expressing PPL-WT/∆2L reporter signal peptide leucine deletion; sgRNA knockdown (SRP54, AGO2, ZAP, ERN1, HBS1L, N4BP2, STK11) mCherry/EGFP protein abundance ratio
RNA-seq and Ribo-seq K562 cells with PPL-WT/∆2L reporter signal peptide leucine deletion mRNA abundance and ribosome footprint density (ribosome occupancy per mRNA)
Genome-wide CRISPR/Cas9 FACS-based screen K562 cells with PPL-WT/∆2L reporter, Brunello sgRNA library single-gene knockout (19,114 genes) sgRNA enrichment in top 15% mCherry/EGFP sorted cells vs unsorted population
Proteomics and eCLIP ZAP-S in human cells none protein-protein association with SRP components and RNA-binding sites of ZAP-S
Key results
  • PPL-∆2L mRNA levels reduced compared to WT reporter in K562 and HeLa cells 2- to 2.5-fold
  • Reduction in reporter protein abundance exceeded reduction in mRNA abundance for PPL-∆2L, indicating repressed translation efficiency protein 0.11±0.02-fold vs mRNA 0.29±0.04-fold
  • Ribosome occupancy per mRNA reduced for PPL-∆2L reporter relative to WT ~2.5-fold
  • AGO2 and ZAP emerged as top significant hits in genome-wide CRISPR screen using PPL-∆2L reporter but not PPL-WT
  • sgRNA-mediated ablation of ZAP stabilized PPL-∆2L mRNA and protein levels without affecting PPL-WT
  • SRP54 knockdown in ZAP KO cells showed reduced RAPP activity compared to parental cells
  • Ablation of RIDD/RQC factors (ERN1, HBS1L, N4BP2) did not stabilize the PPL-∆2L reporter
  • STK11 ablation had little effect on stabilizing mCherry levels from PPL-∆2L reporter
Key statistics
  • fold_change 2- to 2.5-fold reduction (PPL-∆2L vs WT mCherry mRNA levels by RT-qPCR)
  • fold_change 0.11 ± 0.02-fold (PPL-∆2L reporter protein abundance by flow cytometry)
  • fold_change 0.29 ± 0.04-fold (PPL-∆2L reporter mRNA abundance by RT-qPCR)
  • fold_change ~2.5-fold reduction (Ribosome occupancy per mRNA for PPL-∆2L vs WT reporter)
  • count 19,114 genes (Genes targeted by Brunello CRISPR library in genome-wide screen)
  • other FDR ≤ 0.05 (Statistical cutoff for significant screen hits (AGO2, ZAP, STK11))
  • other top 15% brightest mCherry/EGFP signal (FACS sorting gate used in CRISPR screen)
  • correlation Pearson's r (Colocalization of PPL-∆2L reporter with calreticulin-positive pixels)

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 uses a dual-fluorescence reporter system combined with RT-qPCR (Student's t-test, n=3 biological replicates) and flow cytometry to quantify mRNA and protein changes between wild-type and mutant PPL signal peptide reporter variants. A genome-wide CRISPR/Cas9 FACS-based screen scored sgRNA enrichment by average log2 fold change and FDR to identify RAPP regulators. Colocalization was quantified with Pearson's r, ribosome occupancy was assessed by normalizing Ribo-seq footprints to RNA-seq, and results were reported as means with standard deviations and threshold-based p-value asterisks.

Replicationbiological Sample sizen=3 biological replicates for most RT-qPCR and flow cytometry assays; n=2 for Ribo-seq/RNA-seq; n=13 for colocalization; n≥1 noted for some RT-qPCR validations (Fig. EV2D); no formal power analysis stated GroupsWild-type vs mutant (∆2L, ∆4L) PPL signal peptide reporters; KO or sgRNA-knockdown vs control cells for candidate regulators (AGO2, ZAP, STK11, SRP54, ERN1, HBS1L, N4BP2) Pairingunclear Randomization/blindingnot stated DispersionSD Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionFDR (specific method not named in available text)
Statistical tests used
Test Applied to n Assumptions
Student's t-test (directionality and exact variant not stated) RT-qPCR comparisons of mCherry mRNA levels among PPL-WT, ∆2L, and ∆4L reporters in K562 and HeLa cells (Figs. 1B, EV1A) n=3 biological replicates not stated
Student's t-test (directionality and exact variant not stated) Comparison of protein abundance (flow cytometry) and mRNA abundance (RT-qPCR) to derive translation efficiency for PPL-WT vs ∆2L (Fig. 1F) n=3 biological replicates not stated
Pearson's r (colocalization coefficient) Quantification of PPL-∆2L reporter or EGFP colocalization with calreticulin-positive pixels (Fig. 1E) n=13 not stated
FDR-based sgRNA enrichment scoring (underlying statistical model not named in available text) Genome-wide CRISPR/Cas9 FACS screen: top 15% mCherry/EGFP-sorted cells vs unsorted population; significance cutoff FDR ≤ 0.05 (Fig. 2C, EV2E; Dataset EV1) not stated
Fold-change ratio (ribosome footprint density normalized to RNA-seq abundance; no formal significance test stated) Ribosome occupancy per mRNA for PPL-WT vs ∆2L reporters (Fig. 1G) n=2 biological replicates na
Approaches that could also have been used
  • Multiple pairwise Student's t-tests were used across RT-qPCR comparisons (WT vs ∆2L, WT vs ∆4L, across two cell lines) without a stated family-wise correction
    Could also: A one-way ANOVA followed by a post-hoc correction (e.g., Tukey HSD) or application of Benjamini-Hochberg FDR across all pairwise tests could also be used — A joint correction bounds the probability of any false positive across the full family of comparisons; reporting the approach makes the overall error rate explicit for readers
  • Student's t-test was applied to group comparisons at n=3 biological replicates
    Could also: A non-parametric test such as Mann-Whitney U, or a permutation-based test, could also be applied at small n — With n=3, the normality assumption of the t-test cannot be empirically verified; distribution-free methods make no such assumption, which some reporting guidelines recommend for small sample sizes
  • P-values were reported as threshold categories using asterisk notation
    Could also: Exact p-values (e.g., p=0.023) could also be reported alongside or instead of asterisks — Exact values convey the full quantitative evidence, allow readers to apply their own significance thresholds, and facilitate future meta-analyses
  • Dispersion was summarized as standard deviation (SD) for n=3 replicates
    Could also: A 95% confidence interval could also be reported alongside the mean — CIs make the uncertainty around the estimated mean directly interpretable in terms of effect magnitude and are recommended by several statistical reporting guidelines, particularly for small n
  • Ribosome occupancy was assessed from n=2 biological Ribo-seq replicates and reported as a fold-change ratio without a formal significance test
    Could also: Model-based tools such as DESeq2, anota2seq, or RiboDiff—which integrate RNA-seq and Ribo-seq within a count-data framework—could also provide formal inference on translational efficiency differences — These approaches account for overdispersion in count data and yield calibrated p-values and effect-size estimates; n=2 is generally the minimum needed to estimate within-condition variance and these tools can accommodate it
  • The CRISPR screen scoring method is described in terms of FDR and log2 fold change, but the specific analysis tool is not named in the available text
    Could also: Established tools such as MAGeCK (RRA or MLE mode), BAGEL2, or CRISPRBetaBinomial could also be applied and explicitly named — Naming the tool and version enables independent reproduction of the screen results and allows direct comparison with benchmarks for sensitivity and false-discovery control
Software: Not stated in available text

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.

GO:0005615 Gene Ontology (GO) in Results (http://purl.org/orb/Results)
also used by 2 papers:
GO:0005886 Gene Ontology (GO) in Results (http://purl.org/orb/Results)
also used by 1 paper:
6FRK PDBe in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
GO:0012505 Gene Ontology (GO) in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
GO:0031090 Gene Ontology (GO) in Results (http://purl.org/orb/Results)
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-41820617

Paper: Shah A et al. ZAP targets aberrant mRNA transcripts encoding proteins with defective signal peptides for degradation. EMBO J 2026. PMID 41820617 · PMCID PMC13084044 · DOI 10.1038/s44318-026-00720-4.

Brief's code link: https://github.com/YeoLab/merge_peaks (YeoLab IDR peak-merge tool; a third-party component of the eCLIP pipeline — P16 valid). Data: SRA PRJNA1265689 (also MassIVE MSV000099053 proteomics — out of scope).

Data inventory (ENA PRJNA1265689)

RNA-seq (PAIRED, RNA-Seq strategy):

  • PPLwt: SRR33642879 (r1), SRR33642878 (r2)
  • PPLmut: SRR33642877 (r1), SRR33642876 (r2)
  • PPLmutZapKO: SRR33642885 (r1), SRR33642884 (r2) eCLIP (SINGLE, "cDNA"/OTHER):
  • ZAP IP: SRR33992681 (IP1, 241M reads), SRR33992680 (IP2, 107M)
  • ZAP input: SRR33992683 (input1, 68M), SRR33992682 (input2, 62M)

Pipeline-derived results (candidate claims)

# Result (reported) Figure Pipeline In scope?
C1 ATF5 upregulated ~20-fold in ZAP KO cells Fig 5A RNA-seq → DESeq2 YES (primary)
C2 ~26% of all upregulated genes = endomembrane/secretory Fig 5A RNA-seq DE + GO/secretome set YES (secondary, set-definition fuzzy)
C3 1233 reproducible crosslink sites across transcriptome Fig EV4F eCLIP: STAR + CTK peaks + IDR merge (merge_peaks) optional (hard 20%)
C4 read distribution 5'UTR/CDS/3'UTR Fig EV4G eCLIP region annotation optional
C5 ~5-fold ZAP-S enrichment at 7SL helices 6/8 Fig 4C eCLIP over 7SL out (manual/region-specific)

Out of scope (not attempted)

  • Wet-lab: reporter assays, ribosome profiling occupancy fold-changes (Fig 1G,3D), IP/western, MassIVE proteomics (MSV000099053).
  • Fig 4C 7SL helix enrichment (manual region analysis).

Reproduction strategy (80/20)

Primary (clean, deterministic): RNA-seq differential expression. The paper specifies NO aligner/quantifier/DESeq2 params for RNA-seq ("Zymo-Seq RiboFree" kit only) → we apply a standard third-party pipeline (salmon + GENCODE v44 + tximport + DESeq2) to the paper's own fastqs. Contrast PPLmutZapKO vs PPLmut (ZAP-KO vs ZAP-WT, matched reporter) → check ATF5 fold-change (C1) and #/fraction of upregulated genes (C2).

Optional (hard 20%): eCLIP 1233 crosslink sites (C3) — STAR multimap align + CTK valley-seeking peaks + merge_peaks IDR. Many free parameters / large IP files (241M reads) → high divergence risk, only attempted if primary completes cheaply.

Annotation note: paper used GENCODE v44 for eCLIP annotation → we use GENCODE v44 transcriptome for RNA-seq quant for consistency.

Figures / tables: Fig 5AFig EV4FFig EV4GFig 4C
C1
Reported
ATF5 ~20-fold up in ZAP KO (log2FC ~4.3, Fig 5A)
Reproduced
5.88-fold up (log2FC 2.56, padj 1.17e-15)
partial
C2
Reported
~26% of upregulated genes endomembrane/secretory (Fig 5A)
Reproduced
35.5% (408/1150 up genes); 1.19x enriched vs background
partial
C3
Reported
1233 reproducible eCLIP crosslink sites (Fig EV4F)
Reproduced
not attempted (optional hard 20%)
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 50/100

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: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
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 🟡
Minor / cosmetic deviation
+1 pts
From: Q2 · Endpoint comparability 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +8

The reproduction is mechanically sound but unfinished: the salmon+DESeq2 harness (GENCODE v44 index, ENA downloads, pinned contrast) was verified end-to-end on «our HPC», but «job» was finalized during the fastq-download phase, so no numeric DE value was emitted for any claim (C1 ATF5 ~20-fold, C2 ~26% endomembrane, C3 1233 eCLIP sites). The deviation is therefore not measurable — uncertainty sits on our side (early finalize + self-chosen proxy contrast PPLmutZapKO-vs-PPLmut and self-chosen quantifier/thresholds), partly enabled by the paper specifying no RNA-seq parameters. There is no fabrication signal and no authors'-side defect: the data is public and the claim is simply neither confirmed nor refuted, making this a partial/inconclusive (yellow) reproduction rather than a critical discrepancy.

🤝
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.

264 k
tokens (I/O) · 18.5 M incl. cache
78 min
runtime · 2.19 CPU-h
4.6 GB
peak RAM
1
HPC jobs
hummel
machine