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

RIT1 controls actin dynamics via complex formation with RAC1/CDC42 and PAK1.

PLoS Genetics 14(5):e1007370 · 2018
L1 No computation 2/4
Why this verdict

Part of the results reproduced; minor but material deviations remained.

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 🟡
Input / endpoint not comparable 1:1
+1 pts
From: Q1 · Data identity 🔴
Total score +8
✓ What held up
  • No authors-side cause for any deviation
  • Any deviation was negligible
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
  • 🟡Reported values were not (fully) derivable from the shared data
  • 🟡The central claim did not (fully) hold under reproduction
  • 🟡Overall, the reproduction showed a material discrepancy
Reproduction agent’s raw note

DROP (non_pipeline). PMID 29734338 = Meyer Zum Buschenfelde et al., 'RIT1 controls actin dynamics via complex formation with RAC1/CDC42 and PAK1', PLoS Genet 2018;14(5):e1007370 (PMC5937737). Full-text Methods + Data-availability reviewed: it is a wet-lab molecular/cell-biology study (recombinant GST/His proteins, GST pull-down, co-IP, immunoblot/autoradiography, immunofluorescence microscopy, transwell migration, FACS counting). NO bioinformatic/computational pipeline and NO deposited data or code: data-availability statement is 'All relevant data are within the paper and its Supporting Information files' (no GEO/SRA/ENA/figshare/Zenodo accession, no GitHub/GitLab repo). The only computational steps are ImageJ densitometry/manual cell counting and GraphPad Prism 7 statistics applied to wet-lab measurements; the raw images/blots/per-replicate values are not deposited, so nothing is pipeline-reproducible. Apparent term hits (BLAST/align/genom/pipeline) are false positives (neuroBLASToma, mALIGNancies, genomic-location cloning, and a cited reference title). No «our HPC» compute submitted (correctly: no compute-bearing analysis exists). Nothing attempted to reproduce; honest drop, not a failure to run. See scope.md + claims.tsv (4 wet-lab results recorded for audit, all out-of-scope).

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
    assessed: 2026-06-18 ⛓ aaecf2c77d3f
✎ 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 2026-06-18
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

RIT1, a RAS-family GTPase mutated in Noonan syndrome, may act through signaling routes beyond RAF-MEK-ERK and PI3K-AKT, specifically via cytoskeleton-regulating effectors such as PAK1 and the RHO GTPases RAC1/CDC42, and disease-causing RIT1 mutations may alter this novel signaling node.

Core claims
  • PAK1 is a novel direct effector of RIT1 finding
  • RIT1 directly interacts with the RHO GTPases CDC42 and RAC1 finding
  • RIT1 interactions with PAK1, CDC42 and RAC1 are independent of the guanine nucleotide (GDP/GTP) bound to RIT1 mechanism
  • NS-associated RIT1 mutations enhance protein-protein interaction with PAK1, CDC42 or RAC1 and uncouple complex formation from serum/growth factor regulation finding
  • The RIT1-PAK1 complex regulates cytoskeletal rearrangements, causing dissolution of stress fibers and reduction of mature paxillin-containing focal adhesions in COS7 cells, an effect blocked by dominant-negative CDC42/RAC1 or kinase-dead PAK1 finding
  • RIT1 wildtype and NS-associated variants enhance cell motility in a transwell migration assay finding
  • NS-associated RIT1 mutants increase ERK1/2 phosphorylation compared with wildtype finding
  • RIT1 interaction with PAK1 is stimulated by serum factors finding
Experimental setups
Assay System Perturbation Readout Platform
Western blot / immunoblotting (pERK1/2, ERK1/2) HEK293T cells overexpression of RIT1 WT/mutants (K23N, G31R, A57G, F82L, M90V, G95A) phospho-ERK1/2 normalized to total ERK1/2 upon serum starvation/stimulation
Western blot / immunoblotting (pAKT S473/T308) HEK293T cells overexpression of RIT1 WT/mutants, EGF stimulation phospho-AKT levels
GST pull-down (GST-RALGDS[RA], GST-PLCE1[RA], GST-PIK3CA[RBD], GST-PAK[CRIB]) HEK293T cells serum starvation (0.1%), EGF stimulation, or full serum (10%) amount of co-precipitated HA-RIT1
GST pull-down (GST-PAK[CRIB]) HEK293T cells RIT1 WT vs NS-associated mutants (K23N, G31R, A57G, F82L, M90V, G95A) relative amount of co-precipitated HA-RIT1
Co-immunoprecipitation (anti-PAK1) HEK293T cells RIT1 WT vs mutants, serum-starved or full serum co-precipitated HA-RIT1 normalized to IP'd PAK1 and total lysate
Co-immunoprecipitation (endogenous PAK1/RIT1) COS7 cells RIT1 WT vs p.G95A interaction between endogenous PAK1 and RIT1
GST/GFP pull-down (GFP-PAK4) HEK293T cells RIT1 WT vs p.G95A presence of HA-RIT1 in GFP-PAK4 precipitates (group II PAK specificity test)
In-vitro binding assay with purified recombinant proteins purified His-tagged RIT1 and GST-PAK[CRIB] RIT1 loaded with GTPγS vs GDP; A57G and F82L mutants direct binding/affinity of RIT1 to PAK[CRIB]
Key results
  • All six NS-associated RIT1 mutants induced elevated and prolonged ERK1/2 phosphorylation upon serum stimulation compared with WT; significant for p.F82L (5 min), p.G95A (30 min), p.M90V (5,15,30 min)
  • RIT1 p.G95A significantly stimulated ERK1/2 phosphorylation under serum-deprived conditions, unlike WT or other mutants
  • Under steady-state (10% serum), RIT1 p.K23N showed statistically significant increased ERK1/2 phosphorylation vs WT
  • HA-RIT1 co-precipitation with GST-PIK3CA[RBD] was significantly increased in cells cultured with serum factors vs serum-starved cells
  • HA-RIT1 co-precipitation with GST-PAK[CRIB] was significantly increased in cells cultured with serum (10%) vs serum-starved (0.1%) cells
  • RIT1 p.K23N and p.G95A significantly increased co-precipitation with PAK[CRIB] compared with WT; other mutants showed a consistent but non-significant trend
  • Co-IPed HA-RIT1 p.M90V with endogenous PAK1 was significantly increased vs WT in serum-deprived cells; p.F82L and p.G95A showed a similar trend
  • His-tagged RIT1 directly bound GST-PAK[CRIB] in vitro in the nanomolar affinity range, slightly increased when RIT1 was GTPγS-bound vs GDP-bound nanomolar affinity
Key statistics
  • pvalue P < 0.05 (ERK1/2 phosphorylation increase for RIT1 p.F82L (5 min) vs WT after serum stimulation)
  • pvalue P < 0.01 (ERK1/2 phosphorylation increase for RIT1 p.G95A (30 min) vs WT after serum stimulation)
  • pvalue one-way ANOVA P < 0.01 (steady-state ERK1/2 phosphorylation among RIT1 p.K23N, p.G31R, p.M90V vs WT)
  • pvalue one-way ANOVA P < 0.01 (co-precipitated RIT1 with PIK3CA[RBD]/PAK[CRIB] under serum vs serum-starved conditions)
  • pvalue one-way ANOVA P < 0.01 (co-precipitation of RIT1 WT and mutants with PAK[CRIB])
  • pvalue P < 0.05 (co-IP of RIT1 mutants (incl. M90V) with endogenous PAK1 under serum starvation)
  • other ≥5% (frequency of RIT1 mutations among Noonan syndrome cases)
  • count 3 independent experiments (replicate number for most immunoblotting/densitometry quantifications)

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 cell-biology study used transient transfection of heterologous expression systems (HEK293T, COS7) and in-vitro binding assays to characterize RIT1–PAK1/RAC1/CDC42 interactions and downstream cytoskeletal effects. Quantitative comparisons of immunoblot densitometry data relied primarily on one-way ANOVA followed by Bonferroni-corrected post-hoc t-tests, with some pairwise comparisons made directly by unpaired t-tests. Results were reported as mean ± SD of 2–4 independent experiments, with significance thresholds at P < 0.05, < 0.01, and < 0.001; exact p-values and effect sizes were not reported.

Replicationbiological Sample sizeStated per figure as number of independent experiments (range n=2–4); no formal power calculation mentioned GroupsRIT1 wildtype vs. six NS-associated mutants (p.K23N, p.G31R, p.A57G, p.F82L, p.M90V, p.G95A) across serum conditions (0.1%, 10%, EGF stimulation) Pairingunpaired Randomization/blindingnot stated DispersionSD Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionBonferroni correction applied as post-hoc step following one-way ANOVA; no correction stated for standalone unpaired t-tests (Fig 1B, 3B, 3C)
Statistical tests used
Test Applied to n Assumptions
Unpaired two-sample t-test Fig 1B – ERK1/2 phosphorylation at each serum-stimulation timepoint (0, 5, 15, 30 min), mutants vs. wildtype 3 independent experiments not stated
One-way ANOVA with post-hoc unpaired t-tests and Bonferroni correction Fig 1C – steady-state ERK1/2 phosphorylation across RIT1 wildtype and mutant groups 3 independent experiments not stated
One-way ANOVA with post-hoc unpaired t-tests and Bonferroni correction Fig 2B – relative amount of co-precipitated RIT1 across pull-down conditions (RALGDS, PLCE1, PIK3CA, PAK[CRIB]) 4 independent experiments not stated
One-way ANOVA with post-hoc unpaired t-tests and Bonferroni correction Fig 3A – relative co-precipitation of RIT1 wildtype and mutants with GST-PAK[CRIB] 3 independent experiments not stated
Unpaired t-test Fig 3B – co-IP of HA-RIT1 wildtype vs. p.G95A with endogenous PAK1 under 0.1% and 10% serum 3 independent experiments not stated
Unpaired t-test Fig 3C – co-IP of HA-RIT1 wildtype vs. multiple mutants with endogenous PAK1 under serum deprivation 2 independent experiments not stated
Approaches that could also have been used
  • In Fig 1B, separate unpaired t-tests were applied at each of four timepoints for each mutant, generating many simultaneous comparisons without a stated family-wise correction
    Could also: A two-way repeated-measures ANOVA (factors: construct × time) with a single post-hoc correction (e.g., Tukey HSD or Dunnett vs. wildtype) could also have been used — A factorial model would simultaneously test main effects and the interaction of construct and time, and a single post-hoc correction step would directly control the family-wise error rate across all timepoint-by-group comparisons
  • Several key comparisons (Fig 3B, 3C) were made with n=2–3 independent experiments using unpaired t-tests that assume approximate normality
    Could also: Non-parametric alternatives such as the Mann-Whitney U (Wilcoxon rank-sum) test could also be applied — With very small n, the central-limit theorem provides limited assurance of normality; non-parametric tests make no distributional assumptions, though they sacrifice some power when the normality assumption is in fact satisfied
  • Dispersion was reported as SD throughout
    Could also: 95% confidence intervals around the mean could also be reported alongside or instead of SD — CIs convey both the spread of the data and the precision of the mean estimate, and are increasingly requested by journals as they facilitate meta-analytic combination of results
  • Statistical significance was communicated via asterisk tiers (P < 0.05, < 0.01, < 0.001) rather than exact p-values
    Could also: Reporting exact p-values (e.g., P = 0.023) could also be used — Exact values let readers assess evidence strength on a continuous scale, support replication efforts, and are required for many meta-analyses
  • No effect-size metrics were reported alongside the p-values
    Could also: Standardized effect sizes such as Cohen's d, or unstandardized fold-changes with 95% CI, could also be reported — Effect sizes decouple the magnitude of a difference from sample size, providing additional context especially when n is small and power is limited
  • Normalization divided each replicate by the mean of the wildtype samples within that experiment to conserve relative variance (citing ref 79)
    Could also: Linear mixed-effects models treating experiment as a random effect could also account for between-experiment variability while analyzing absolute densitometry values — Mixed models explicitly partition within- vs. between-experiment variance and can be more powerful when the number of replicates per experiment varies, though they require more observations to estimate random effects reliably
Software: Not stated

What was reproduced

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

Scope — pmid-29734338

Title: RIT1 controls actin dynamics via complex formation with RAC1/CDC42 and PAK1. Authors: Meyer Zum Büschenfelde U, Brandenstein LI, von Elsner L, Flato K, Holling T, Zenker M, Rosenberger G, Kutsche K. Venue: PLoS Genetics 2018 May 7;14(5):e1007370. DOI: 10.1371/journal.pgen.1007370. PMCID: PMC5937737. Data Availability (verbatim): "All relevant data are within the paper and its Supporting Information files."

Verdict: DROP — non_pipeline

This is a wet-lab molecular / cell-biology study. After reading the full text (PMC5937737) Methods + Data-availability statement, there is no bioinformatic / computational pipeline to reproduce, and no deposited data or analysis-code artifact (no GEO/SRA/ENA/ArrayExpress/figshare/Zenodo accession, no GitHub/GitLab repository).

What the paper actually does (all out of scope = wet-lab / manual)

  • Heterologous expression in COS7 / Flp-In 293 / HEK293 cells; purified recombinant GST-/His-tagged proteins.
  • Direct protein–protein interaction by GST pull-down, co-immunoprecipitation, immunoblotting (autoradiography).
  • ERK1/2 phosphorylation immunoblots.
  • Immunofluorescence microscopy of stress fibers / paxillin-positive focal adhesions.
  • Transwell migration / invasion assays; FACS-based cell counting (FACSCalibur + CellQuestPro + AccuCheck beads).

The only "computational" steps present — still NOT a reproducible pipeline

Tool Use in paper Why out of scope
ImageJ (NIH) densitometry of autoradiographs; counting paxillin-positive structures (≥30 cells/dataset), morphology scoring (≥50 cells/dataset) manual/semi-manual quantification of wet-lab images; raw images/blots are NOT deposited — only summary values appear in figures/SI, so the quantification cannot be re-run
GraphPad Prism 7 (InStat) mean ± SD; Student's t-test (pairwise), one-way ANOVA + Bonferroni post-hoc standard statistics applied to the wet-lab measurements above; not a bioinformatic pipeline, and the per-cell/per-blot input values are not provided as machine-readable data

Disambiguation of full-text term hits (false positives)

  • "BLAST" → only "neuroblastoma" / "lymphoblastic".
  • "align" → only "malignancies".
  • "genom(ic)" → "genomic location" (Flp-In integration site, a cloning method) + reference titles.
  • "pipeline" → appears only inside the title of a cited reference ("...Next-Generation Sequencing Diagnostic Pipeline of RASopathies", Sci Rep 2018), not this paper's own methods.
  • RIT1 variants studied (p.K23N, p.A57G, p.G31R, p.F82L, p.M90V, p.G95A) are known Noonan-syndrome alleles introduced by site-directed mutagenesis — no variant-calling / sequencing pipeline.

In-scope pipeline-derived results

None. No reported value in this paper is produced by a bioinformatic pipeline operating on a deposited dataset. Nothing is attempted on «our HPC» (correctly — there is no compute-bearing analysis to run).

Datasets relied upon

None external. No accession exists; all data are figure panels + Supporting Information within the article itself (CC BY). data/dataset_profile.json records this (no external deposit).

Figures / tables: FigsFig 7

No individual results have been recorded for this entry yet.

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 🟡
Input / endpoint not comparable 1:1
+1 pts
From: Q1 · Data identity 🔴
Total score +8

This is a correctly identified non_pipeline drop: Meyer Zum Büschenfelde et al. 2018 (PLoS Genet, PMC5937737) is a wet-lab molecular/cell-biology study (GST pull-down, co-IP, immunoblot, immunofluorescence, transwell migration) with no bioinformatic pipeline and no deposited data/code accession. Nothing was attempted to reproduce because there is nothing pipeline-reproducible — the only 'computation' is interactive ImageJ densitometry and GraphPad Prism statistics on undeposited raw images. The inability to compare values is purely a data-availability/scope matter (q1/q2 red), not an authors' defect or fabrication concern, so q5/q7/q8 stay yellow rather than red. Severity is non-applicable (no deviation measured).

🤝
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.

39.2 k
tokens (I/O) · 1.6 M incl. cache
15 min
runtime
Per-job HPC accounting not captured for this run — the runtime shown is the reproduction’s measured wall-clock time.