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

CDKL1 variants affecting ciliary formation predispose to thoracic aortic aneurysm and dissection.

· 2025
PubMed 41056017 ↗ pmid-41056017
L1 67/100 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: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Total score +4
✓ 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
What did not (or only partly)
  • 🟡The central claim did not (fully) hold under reproduction
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
67/100
Reproducibility score
0.4 SD below mean
vs. all fields · 1173 studies
🎯 Scores higher than 29% of all assessed papers rank 795 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 one pipeline-derived result on PUBLIC data. The paper (JCI 10.1172/JCI186287) is mostly clinical-genetics + wet-lab; its single clearly reproducible computational claim is Fig 3B, the CDKL1 single-nucleus expression reanalysis of two public aortic snRNA-seq datasets (GSE207784 Chou 2022, GSE165824 Pirruccello 2022) merged and Harmony-integrated. R1 reproduced 1:1 (EXACT, provisional): both dataset N's match exactly (71,689 + 54,092 nuclei), CDKL1 is low+ubiquitous across all cell types (global 5.33% expressing), and the joint Harmony-integrated ranking puts the three paper-named cell types (lymphatic endothelial, neuronal, VSMC subtype II) at the very top -- and the pattern replicates independently in BOTH cohorts. R2 (PXD048044 CDKL1 AP-MS) profiled: deposit + .msf search confirmed real (GFP bait #1, 16,885 proteins) but PARTIAL submission means the specific interactor list isn't recoverable 1:1; full re-search not attempted. NOT attempted (out of scope): the 6/323 WES/panel variant calls (restricted human data) and all wet-lab assays. Verdict: strong partial -- the headline reproducible pipeline result is an exact match; remaining claims are either restricted-data or wet-lab.

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 67
    assessed: 2026-06-19 ⛓ 45a5cb4f61b0
✎ 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
2026-06-19
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 study tests whether previously unreported genetic variants, specifically in CDKL1, contribute to the etiology of thoracic aortic aneurysm and dissection (TAAD) given that current known genes explain only ~30% of cases.

Core claims
  • Heterozygous CDKL1 missense variants (Cys143Arg, Ser206Leu, Thr135Met) were identified in 6 patients from 3 families with TAAD spectrum disorders finding
  • CDKL1 encodes a protein kinase involved in ciliary biology mechanism
  • CDKL1 amino acid substitutions are predicted to affect catalytic activity or protein binding properties finding
  • CDKL1 is expressed in vascular smooth muscle cells of normal and diseased human aortic wall tissue finding
  • Cdkl1 knockdown and transient knockout in zebrafish cause intersomitic vessel malformations and aortic dilation finding
  • Wild-type but not mutant (Cys143Arg, Ser206Leu) human CDKL1 RNA rescues zebrafish ISV malformations finding
  • CDKL1 variants alter kinase function and protein-protein binding, particularly with ciliary transport molecules mechanism
  • CDKL1 variants interfere with cilia formation/length, CDKL1 localization, and p38 MAPK and Vegf signaling mechanism
Experimental setups
Assay System Perturbation Readout Platform
exome sequencing human patients (family 1, 3 affected individuals) none candidate disease-associated sequence variants
gene panel sequencing cohort of 320 unrelated individuals with TAAD/connective tissue disorders none detection of CDKL1 and other candidate gene variants
structural bioinformatics / molecular modeling CDKL1 protein structure Cys143Arg, Ser206Leu, Thr135Met substitutions predicted changes in intramolecular interactions, ATP/Mg2+ binding, surface hydrophobicity
immunohistochemistry human aortic wall tissue (3 aneurysm patients, 2 controls) none CDKL1 protein expression/localization relative to ACTA2 and CD31 markers
single-cell RNA-seq (public dataset analysis) human aortic wall cells none CDKL1 expression across aortic cell types/VSMC subtypes
whole-mount in situ hybridization zebrafish embryos none cdkl1 expression pattern in ciliated tissues
morpholino knockdown zebrafish embryos Cdkl1 knockdown (ATG MO, splice MO) with/without human CDKL1 WT or mutant RNA coinjection intersomitic vessel malformation and dorsal aortic dilation
CRISPR/Cas9 transient gene editing zebrafish embryos (crispants) Cdkl1 knockout with/without human CDKL1 WT or mutant RNA coinjection intersomitic vessel malformation and dorsal aortic dilation
Key results
  • 3 distinct heterozygous CDKL1 variants identified in 6 patients from 3 families with TAAD spectrum disorders
  • CDKL1 p.(Ser206Leu) predicted deleterious by all 5 tested pathogenicity classifiers; p.(Thr135Met) by 4 of 5
  • CDKL1 knockdown (MO) and crispant knockout caused ISV malformation, shortening, and breakdown in zebrafish
  • Splice-MO knockdown and CRISPR/Cas9 knockout of Cdkl1 caused significant dorsal aortic dilation
  • Coinjection of human CDKL1 WT RNA partially but significantly rescued ISV malformations from knockdown/knockout
  • Coinjection of CDKL1 Cys143Arg or Ser206Leu RNA failed to rescue ISV phenotype
  • CDKL1 variants altered kinase function, profiling data, and protein-protein binding, especially with ciliary transport molecules
  • CDKL1 variant expression interfered with cilia formation/length, CDKL1 localization, and p38 MAPK and Vegf signaling
Key statistics
  • count 323 patients (total patients undergoing exome and gene panel sequencing)
  • count 320 individuals (cohort screened by gene panel sequencing for CDKL1 variants)
  • count 6 patients from 3 families (carriers of heterozygous CDKL1 variants)
  • mean 38.5 ± 12.4 years (mean age at first clinical manifestation)
  • other allele frequency 0.00001831 (gnomAD v2.1.1 controls) (CDKL1 c.617C>T p.(Ser206Leu) population frequency)
  • other allele frequency 0.00004570 (gnomAD v2.1.1 controls) (CDKL1 c.404C>T p.(Thr135Met) population frequency)
  • other allele count 0 in gnomAD v2.1.1 (CDKL1 c.427T>C p.(Cys143Arg) population frequency)
  • other >25 intramolecular interaction changes (predicted structural effect of p.Cys143Arg substitution)

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 combines a human genetics case series (exome sequencing in one family, then gene panel sequencing in a cohort of 320 additional individuals with TAAD spectrum/connective tissue disorders) with in silico structural/conservation modeling, immunohistochemistry, reanalysis of public single-cell RNA-seq datasets, and functional validation in zebrafish (morpholino knockdown, CRISPR/Cas9 crispants, and mRNA rescue experiments). Results are reported largely descriptively (variant allele frequencies, in silico pathogenicity scores, expression patterns), with statistical significance noted qualitatively (e.g., 'significantly rescued', 'significant aortic dilation') for the zebrafish phenotyping experiments. The specific statistical tests, exact p-values, group sample sizes, and software used for these comparisons are not stated in the main-text passages provided (statistical methods are referenced as residing in Supplemental Methods/Data, not included here).

Replicationunclear Sample sizeNot described in the provided text for the zebrafish experiments (no per-group embryo counts or power justification stated); the patient cohort size is given (323 sequenced patients; 320-individual panel cohort plus 3-individual family exome cohort) GroupsCDKL1 wild-type vs. variant (p.Cys143Arg, p.Ser206Leu) mRNA rescue in zebrafish; Cdkl1 morpholino/CRISPR knockdown or knockout vs. control-injected embryos; CDKL1 variant carriers vs. population reference databases (gnomAD) Pairingunclear Randomization/blindingnot stated Dispersionunclear Exact p-valuesno
Statistical tests used
Test Applied to n Assumptions
not stated (described only as 'significantly rescued') Rescue of ISV malformation by CDKL1 wild-type vs. variant (Cys143Arg, Ser206Leu) RNA coinjection after MO knockdown (Figure 4C) not stated
not stated (described only as 'significantly rescued') Rescue of ISV malformation by CDKL1 wild-type vs. variant RNA coinjection after CRISPR/Cas9 crispant knockdown (Figure 4E) not stated
not stated (described only as 'significant aortic dilation') Dorsal aortic diameter, Cdkl1 splice-blocking MO vs. control embryos (Figure 4, F and G) not stated
not stated (described only as 'significantly dilated') Dorsal aortic diameter, CRISPR/Cas9 Cdkl1 knockout vs. Cas9-only injected embryos (Figure 4H) not stated
in silico pathogenicity prediction (5 classifiers), used as a prioritization/classification criterion rather than an inferential statistical test Variant prioritization for CDKL1 and other candidate variants (Table 2) na
Approaches that could also have been used
  • Patient age at first manifestation is summarized as mean ± a dispersion value without specifying SD or SEM.
    Could also: Reporting the dispersion measure explicitly (SD for describing spread, SEM for describing precision of the mean), or using median with IQR/range given the small cohort (n=6) — Explicit labeling and/or a median-based summary can better convey variability in a small, potentially skewed patient sample.
  • Zebrafish rescue and phenotype comparisons are described as 'significant' without naming a specific statistical test.
    Could also: Reporting the specific test used (e.g., unpaired two-tailed t-test, Mann-Whitney U, or one-way ANOVA with post-hoc comparisons) along with exact p-values — Naming the test and giving exact p-values lets readers evaluate the assumptions made (e.g., normality) and the precision of the significance claim.
  • Multiple CDKL1 RNA conditions (wild-type, Cys143Arg, Ser206Leu, injection-alone) appear to be compared against a knockdown/knockout baseline in the same rescue experiment.
    Could also: A one-way ANOVA (or Kruskal-Wallis) across all conditions with a post-hoc correction (e.g., Dunnett's test comparing each variant to the knockdown-only control, or Tukey HSD for all pairwise comparisons) — An omnibus test with corrected post-hoc comparisons controls the family-wise error rate when several groups are compared against a common reference.
  • Variant pathogenicity was assessed by tallying predictions from 5 individual in silico classifiers (e.g., '4 of 5' or '5 of 5' predicted damaging).
    Could also: An ensemble/meta-predictor score (e.g., REVEL, AlphaMissense, or a calibrated combined likelihood) or reporting continuous scores/thresholds rather than a simple majority count — A calibrated ensemble score can integrate predictor outputs with known accuracy weighting rather than treating each classifier's binary call as equally informative.
  • Association between CDKL1 variants and disease is supported by rarity in gnomAD (allele frequency comparisons) rather than a formal statistical test against controls.
    Could also: A formal case-control burden test (e.g., Fisher's exact test on carrier counts, or a gene-based burden/SKAT-O test) comparing the patient cohort to a matched control population — Formal burden testing can provide an explicit statistical estimate (odds ratio, p-value) for the association beyond frequency comparison alone, which is useful when cohort size allows it.
  • Sample sizes and any power justification for the zebrafish embryo experiments are not stated in the provided text.
    Could also: Reporting the number of embryos per condition/replicate and, where feasible, an a priori power calculation — Stating n per group and a power rationale allows readers to judge whether the study was adequately powered to detect the reported phenotypic differences.

What was reproduced

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

Scope — pmid-41056017

Paper: Nauth T, et al. "CDKL1 variants affecting ciliary formation predispose to thoracic aortic aneurysm and dissection." J Clin Invest 2025. DOI 10.1172/JCI186287. PMC12646653.

What kind of paper

Primarily a wet-lab / clinical-genetics study. Heterozygous CDKL1 variants found in 6/323 TAAD patients (3 families); functional follow-up in zebrafish, in-vitro kinase assays, cell-based cilia assays, phospho-profiling (STK-PamChip), LC-MS/MS substrate ID, immunohistology. Only a small slice is bioinformatic-pipeline-derived.

Results classified

# Reported result Method In scope? Why
R1 Fig 3B — CDKL1 single-cell expression across aortic cell types: "low but ubiquitous expression across all cell types … slightly enriched in lymphatic endothelial cells, neuronal cells, and VSMC subtype II". Reanalysis of 2 public human aortic snRNA-seq datasets (refs 28,29), merged + integrated with Harmony, UMAP feature plot of CDKL1. scRNA/snRNA-seq reanalysis pipeline (Seurat + Harmony) on public data IN SCOPE — PRIMARY Public data (GSE207784, GSE165824), standard reproducible pipeline. This is the one clearly pipeline-derived result on public inputs.
R2 CDKL1 substrate / phospho-interactome via LC-MS/MS; proteomics deposited as PXD048044 (PRIDE/ProteomeXchange). LC-MS/MS proteomics (MaxQuant-class) SECONDARY / partial Data are public (PXD048044), but the reported claim is a qualitative substrate/interactor list tied to proprietary processing + co-IP design; reproducing the exact differential list 1:1 is hard. Will profile the deposit; attempt reanalysis only after R1.
R3 STK-PamChip phospho-profiling — "71 peptides passed QC", differential phosphorylation WT vs variants. Proprietary PamGene BioNavigator kinase-array software OUT (borderline) Vendor-locked array + software; raw not deposited as a reusable pipeline input. Not a standard open pipeline.
R4 Exome + gene-panel sequencing of 323 patients → CDKL1 variant calls. WES/panel variant calling OUT — data restricted "Due to legal and ethical considerations, human sequencing data … will not be made publicly available." No accession. Cannot reproduce.
R5 Zebrafish ISV malformation / aortic dilation rates; kinase activity assays; cilia length; p38/VEGF signaling; IHC. Wet-lab OUT Not computational.

In-scope plan

  • R1 (primary, 80% floor): Download GSE207784 (Chou 2022, snRNA-seq aneurysmal human aorta, 13 samples / 71,689 nuclei) and GSE165824 (Pirruccello 2022, snRNA-seq thoracic aorta, 3 indiv / 54,092 nuclei) on «our HPC» front node → «infra». Build Seurat/Harmony env. Filter, annotate, merge, Harmony-integrate, UMAP, plot CDKL1 feature/expression across cell types. Compare cell-type enrichment pattern to the paper's claim.
  • R2 (stretch): Profile PXD048044; attempt re-quantification if feasible.

Datasets to profile (dataset_profile.json)

  • GSE207784 — Chou 2022, ATVB — snRNA-seq human aorta (aneurysm + control)
  • GSE165824 — Pirruccello 2022, Nat Genet (also Broad SCP1265) — snRNA-seq thoracic aorta
  • PXD048044 — PRIDE — LC-MS/MS proteomics (CDKL1 interactome/substrates)

Out of scope (not attempted), explicitly

  • Human WES/panel variant calls (R4) — restricted data, no accession.
  • All wet-lab assays (R3, R5).
Figures / tables: Fig 3B
R1
Reported
Fig 3B: CDKL1 shows low but ubiquitous expression across all human aortic cell types, slightly enriched in lymphatic endothelial cells, neuronal cells, and VSMC subtype II (reanalysis of GSE207784 + GSE165824, merged + Harmony-integrated).
Reproduced
CDKL1 detected (pct>0) in EVERY annotated cell type in both datasets; global 5.33% nuclei expressing (mean lognorm 0.10). In the joint Harmony-integrated 125,781-nucleus object the top CDKL1 cell types are exactly the three named: lymphatic endothelial (ranks 1-2, both cohorts; mean 0.51/0.41), neuronal (rank 3; 0.31), VSMC subtype II (rank 4; 0.19, highest of VSMC1/2/3 = 0.085/0.189/0.136). Pattern replicates independently in each dataset.
exact
R2
Reported
CDKL1 interactome/substrates via LC-MS/MS, deposited PXD048044 (PRIDE/ProteomeXchange).
Reproduced
Deposit confirmed + Proteome Discoverer .msf parsed: genuine GFP-Trap AP-MS search (GFP bait = protein #1; 16,885 target proteins; 187,652 PSMs; human DB). Specific reported interactor list not recoverable 1:1 (PARTIAL submission, 'Normalization' .msf has quant fields unpopulated); full re-search of ~10GB raw not attempted.
partial
R4
Reported
Heterozygous CDKL1 variants in 6/323 TAAD patients, 3 families (WES/panel).
Reproduced
NOT ATTEMPTED — human sequencing data explicitly withheld (legal/ethical); out of scope.
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 67/100

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: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Total score +4

The single clearly reproducible computational result on public data — Fig 3B CDKL1 single-nucleus expression (reanalysis of GSE207784 + GSE165824) — reproduced exactly (both Ns matched, three named cell types top the joint Harmony ranking, replicated in both cohorts), with no deviation attributable to our method or the authors. The paper's central conclusion (heterozygous CDKL1 variants in 6/323 TAAD patients) and the R2 AP-MS interactor list could not be verified because the human sequencing data are restricted and PXD048044 is a partial deposit — availability/scope limits on the authors'/data side, not discrepancies or fabrication. Net: a high-quality 1:1 reproduction of the one in-scope claim, but only limited confirmation of the study's headline clinical conclusion.

🤝
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.

212.8 k
tokens (I/O) · 22 M incl. cache
34 min
runtime · 0.27 CPU-h
15 GB
peak RAM
3
HPC jobs
hummel
machine