Shared and unique phosphoproteomics responses in skeletal muscle from exercise models and in hyperammonemic myotubes.
The main results reproduced: recomputed values matched the published ones within tolerance.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
- ✓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
- ✓The central claim held under reproduction
- ✓Overall, the reproduction was clean
- Every checked point held up.
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
1:1 REPRODUCED. The listed code github.com/dusadrian/venn is the third-party venn R package that draws the paper's Venn diagrams (valid P16 target). The in-scope, pipeline-derived result = the set-overlap region counts of differentially expressed phosphoproteins (DEpP) and differentially phosphorylated phosphosites (DPPS). The authors deposited the exact 0/1 membership matrices behind every Venn as supplementary Excel sheets (mmc2/mmc12/mmc15). On «our HPC» compute node n042 (SLURM «job») we freshly downloaded the supp tables to «infra» and recomputed region counts (row-combination counts = exactly the venn zone algorithm). ALL 12 figure Venns (Fig 1A, 4A, 4E, 5A, 5F; ~30 region counts) match the published numbers EXACTLY, in correct labeled order, with marginal totals also matching figure subtitles — zero deviation. NOT attempted (out of scope, stated honestly): upstream MS quantification + significance calls (Proteome Discoverer V2.3 + Perseus 1.5.8.5, licensed Windows GUI, raw .raw only in PXD031372) which DEFINE set membership; pathway enrichment (IPA/DAVID/g:Profiler/STRING/NetworKIN); wet-lab validation. Data note: the BRIEF's accession geo:GSE171642 is ATAC-seq from a sibling deposit, NOT this paper's phosphoproteomics (correct deposit = PRIDE PXD031372); all three profiled. Verdict provisional pending 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.
-
v1 current initial assessment Score 100assessed: 2026-06-22 ⛓ 2bf5c51edfba
✎ 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.
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
- 2026-06-29
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-22no 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: sonnetThe paper tests whether skeletal muscle phosphoproteomic responses to hyperammonemia share molecular pathways (e.g., PKA, calcium, MAPK signaling, protein homeostasis) with responses to exercise, with the goal of identifying preclinical exercise models that best recapitulate human exercise responses.
- ★ Comparative phosphoproteomics of hyperammonemic myotubes and exercise-model muscle identifies shared enriched pathways: PKA, calcium signaling, MAPK signaling, and protein homeostasis. finding
- ★ Hyperammonemic myotubes show distinct temporal phosphorylation patterns, with early (6h) changes in PKA, matrix metalloprotease and integrin signaling, and later (24h) changes in cell cycle control, DNA damage signaling, and PKA signaling. finding
- ★ PKA signaling is the curated pathway enriched in both early (6h) and late (24h) hyperammonemia phosphoproteomics datasets. finding
- ★ Key phosphoproteins were experimentally validated by immunoblot: increased phosphorylation of IKKβ (Ser672) and MST2 (Ser316), and decreased phosphorylation of MCM2 (Ser139) and S6 ribosomal protein (Ser235/236) during hyperammonemia. finding
- ★ A comparative bioinformatics and machine-learning-based approach integrating public and experimentally derived phosphoproteomics data can be used to select preclinical models that recapitulate specific human exercise responses. method
- ★ The HIPPO signaling core kinase Mst2 (Stk3/4) shows increased phosphorylation during hyperammonemia, consistent with a role in muscle atrophy. finding
- Decreased Rps6 phosphorylation during hyperammonemia is consistent with previously reported decreases in protein synthesis. finding
- ★ DPPS in hyperammonemic myotubes cluster into 5 temporal patterns (persistent increase/decrease, late increase/decrease, transient change), each enriched for distinct pathways. finding
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| Phosphoproteomics (mass spectrometry, untargeted) | C2C12 murine myotubes | 10mM ammonium acetate (AmAc), 6h and 24h | Differentially expressed phosphoproteins (DEpP) and phosphorylated phosphosites (DPPS) vs untreated controls | — |
| Immunoblot (Western blot) with densitometry | C2C12 murine myotubes | 10mM ammonium acetate, 6h and 24h | Phosphorylation levels of pIKKβ(S672), pMST2(S316), pMCM2(S139), pS6(S235/236) and total protein | — |
| Comparative bioinformatics / meta-analysis of public phosphoproteomics datasets | Human and mouse skeletal muscle exercise models | exercise | Enriched signaling pathways (PKA, calcium, MAPK, protein homeostasis) | — |
| Protein-protein interaction network analysis | C2C12 myotube phosphoproteomics data | ammonium acetate (6h, 24h) | Most connected differentially phosphorylated proteins / novel signaling cascades | — |
| Cross-omics comparison (ATACseq, RNAseq, proteomics) | Hyperammonemic myotubes, mouse skeletal muscle, human skeletal muscle from cirrhosis patients | ammonia treatment / hyperammonemia (disease) | Overlap of differentially expressed molecules with DEpP in PKA and other pathways | — |
| Functional enrichment analysis (IPA, DAVID, Perseus) | C2C12 myotube phosphoproteomics datasets (6hAmAc, 24hAmAc, shared/unique subsets) | ammonium acetate | Pathway enrichment scores (e.g., PKA, PLK, CDK, HIPPO, HIF1α signaling) | — |
| Hierarchical clustering / dimensionality reduction with feature selection | C2C12 myotube DPPS data | ammonium acetate (6h, 24h) | Temporal clusters of phosphorylation change and their pathway enrichment | — |
- – 448 total DEpP identified in hyperammonemic myotubes: 164 total (75 unique) at 6hAmAc, 373 total (284 unique) at 24hAmAc, 89 shared
- – 617 total DPPS identified: 193 total (108 unique) at 6hAmAc, 509 total (424 unique) at 24hAmAc, 85 shared
- ▼ More DEpP/DPPS were downregulated than upregulated in each treatment group 6hAmAc: 122(74.3%) DOWN vs 49(29.9%) UP; 24hAmAc: 255(68.4%) DOWN vs 158(42.4%) UP
- – Immunoblot validation confirmed increased phosphorylation of IKKβ(S672) and MST2(S316), and decreased phosphorylation of MCM2(S139) and S6(S235/236) in hyperammonemic vs untreated myotubes
- – PKA signaling pathway enriched in both 6hAmAc and 24hAmAc phosphoproteomics datasets
- – 287 DPPS segregated into 5 temporal clusters during hyperammonemia: persistent increase (n=71), persistent decrease (n=135), late increase (n=28), late decrease (n=32), transient change (n=21)
- – No significant correlation between 24hAmAc differentially expressed proteins (DEP) and DEpP; of 19 molecules shared between datasets, 10 showed concordant direction of expression
- – HIF1α signaling identified as significant in the 24hAmAc dataset using non-differentially phosphorylated proteins as background p<0.05
- count 448 DEpP (164 at 6hAmAc, 373 at 24hAmAc, 89 shared) (DEpP in hyperammonemic myotubes vs untreated controls)
- count 617 DPPS (193 at 6hAmAc, 509 at 24hAmAc, 85 shared) (DPPS in hyperammonemic myotubes)
- other 6hAmAc: 122(74.3%) DOWN vs 49(29.9%) UP; 24hAmAc: 255(68.4%) DOWN vs 158(42.4%) UP (Direction of phosphorylation change vs controls)
- pvalue p-adj<0.05 (Student's t-test with Benjamini-Hochberg correction) (Significance cutoff for DEpP/DPPS)
- fold_change log2 ratio>|2.5| and padj<0.05 (IPA significance cutoff for full datasets)
- count 287 DPPS across 5 clusters: 71, 135, 28, 32, 21 (Hierarchical clustering of temporal phosphorylation patterns)
- count 19 shared DEP and DEpP, 10 concordant (Overlap between hyperammonemic proteomics and phosphoproteomics datasets)
- count n=3 biological replicates per group (All myotube phosphoproteomics and immunoblot experiments)
Statistical methods review
Model: sonnetA 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 analyzed phosphoproteomics data from hyperammonemic C2C12 myotubes (6h and 24h ammonium acetate treatment vs untreated controls, n=3 biological replicates per group) using Student's t-tests with Benjamini-Hochberg FDR correction to define differentially expressed phosphoproteins and phosphosites, followed by functional/pathway enrichment analysis using IPA, DAVID, and Perseus with tool-specific significance cutoffs. Selected proteins were validated by immunoblot, with densitometry results reported as mean ± SD and compared by ANOVA with Bonferroni post-hoc testing. These in-house results were also compared descriptively (via overlap/Venn and correlation analyses) with previously published proteomics/transcriptomics datasets and public exercise phosphoproteomics data.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| Student's t-test with Benjamini-Hochberg FDR correction (padj<0.05) | identification of differentially expressed phosphoproteins (DEpP) and phosphosites (DPPS) in 6hAmAc and 24hAmAc myotubes vs untreated controls | n=3 biological replicates (one 24hAmAc replicate excluded as an outlier) | not stated |
| One-way ANOVA with Bonferroni post-hoc analysis | immunoblot densitometry validation of p-IKKβ, p-MST2, p-MCM2, p-S6 across untreated/6hAmAc/24hAmAc myotubes | n=3 biological replicates | not stated |
| Pathway enrichment analysis (IPA), significance cutoff -log(p value) ≥ 1.3 | functional enrichment of DEpP/DPPS datasets | — | not stated |
| Functional enrichment analysis (DAVID), foreground padj<0.05 | functional enrichment of DEpP datasets | — | not stated |
| Perseus 1D annotation enrichment, default BH-FDR>0.02 | functional enrichment of phosphoproteomics datasets | — | not stated |
| Correlation analysis | comparison of expression levels between DEpP and previously published differentially expressed proteins (Figure S2A) and between PKA/PLK pathway components (Figure S5C) | — | not stated |
-
Differential phosphoprotein/phosphosite calls at 6h and 24h were each tested with Student's t-tests and BH-FDR correction applied per timepoint dataset.↳ Could also: A mixed-effects or repeated-measures model with time and treatment as factors (e.g., a time × treatment interaction term) — This would let the significance of temporal change be tested directly within a single model rather than through separate per-timepoint comparisons, and could clarify whether early/late responses differ statistically from one another.
-
Differential expression testing on phosphoproteomics data used conventional per-feature Student's t-tests with n=3 biological replicates per group.↳ Could also: Moderated t-tests with empirical Bayes variance shrinkage (e.g., limma), which are widely used for small-n omics data — Borrowing variance information across many features can stabilize variance estimates when replicate numbers are small, which is a common consideration in high-dimensional omics analyses.
-
Immunoblot validation results were summarized with mean ± SD and significance indicated by p-value threshold categories.↳ Could also: Reporting effect sizes with 95% confidence intervals alongside the p-values — CIs convey the precision and magnitude of an estimated difference, which can be informative to readers in addition to a significance threshold, particularly with small sample sizes.
-
One 24hAmAc phosphoproteomics replicate was excluded from downstream analysis based on outlier status.↳ Could also: Robust or rank-based statistical methods (e.g., Mann-Whitney U, robust regression) that down-weight rather than exclude atypical observations — This approach retains all collected data points while limiting the influence of atypical values, and results with/without the excluded replicate could also be compared as a sensitivity check.
-
Pathway/functional enrichment was performed across three different platforms (IPA, DAVID, Perseus), each using its own default significance cutoff (-log(p)≥1.3, padj<0.05, BH-FDR>0.02, respectively).↳ Could also: Applying a single harmonized enrichment statistic/threshold (e.g., hypergeometric or Fisher's exact test with a consistent FDR cutoff) across all platforms — A common threshold can make it more straightforward to directly compare enrichment strength or overlap of pathways identified between tools.
-
Concordance between the in-house myotube phosphoproteomics data and previously published proteomics/transcriptomics and public exercise datasets was assessed through descriptive overlap (Venn diagrams) and correlation plots.↳ Could also: Formal meta-analytic approaches such as rank aggregation or combined effect-size meta-analysis across datasets — These methods can provide a quantitative statistical summary of concordance across independent datasets, complementing descriptive overlap visualization.
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — pmid-36345342
Paper: Welch et al. 2022, iScience — "Shared and unique phosphoproteomics responses in skeletal muscle from exercise models and in hyperammonemic myotubes." DOI 10.1016/j.isci.2022.105325 · PMID 36345342 · PMCID PMC9636548
Listed code: https://github.com/dusadrian/venn — this is the venn R
package by Adrian Dusa (a general-purpose third-party tool that draws Venn
diagrams for up to 7 sets). It is NOT the authors' own analysis code; per brief
rule P16, applying this existing third-party tool to the paper's own data is an
equally valid reproduction. The paper's central figures (Venn diagrams of
shared/unique phosphosites and phosphoproteins) are produced by this package.
Listed data (brief): geo:GSE171642. This is WRONG / mismatched for the
phosphoproteomics: GSE171642 is the ATAC-seq subseries (9 samples) of a
sibling hyperammonemia paper. The actual mass-spectrometry phosphoproteomics for
THIS paper is PRIDE/ProteomeXchange PXD031372 (18 samples, Orbitrap Fusion
Lumos, Proteome Discoverer V2.3 + Perseus 1.5.8.5). Exercise comparison datasets
are reused public PRIDE/MassIVE deposits (PXD026955, PXD026461, PXD001543,
PXD014322, PXD010452, MSV000086732). GEO GSE171643/4/5 = RNA-seq/proteomics of
the sibling integrated-landscape paper.
What the reported results are produced by (pipeline map)
| Result | Pipeline | In scope? |
|---|---|---|
| Raw MS → identified/quantified phospho-peptides (14453 peptides, 9232 phosphopeptides, 18 samples) | Proteome Discoverer V2.3 (Sequest) + Perseus 1.5.8.5 (label-free, ANOVA/t-test, BH-FDR, imputation) — licensed Windows GUI software | OUT — env_unresolvable on Linux/«our HPC»; raw .raw files only. Not attempted. |
| Differential phosphosite/phosphoprotein lists per condition (DPPS / DEpP) | Perseus statistics (padj<0.05, BH) | OUT (same Perseus GUI step). The outputs of this step ship as Supplementary Tables S1–S28 → used as INPUT below. |
| Venn diagram overlap counts — shared / unique DPPS & DEpP across conditions (Fig 1A: 6h vs 24h hyperammonemia; Fig 4A–B: mouse vs human exercise; Fig 5A: 7-way exercise+hyperammonemia) | venn R package set intersection on the per-condition lists in Suppl. Tables |
IN SCOPE — pure set operations on the paper's shipped lists; reproduce with the listed third-party tool. THIS is the reproduction target. |
| Temporal k-means clusters (Fig 3A: n=71/135/28/32/21 DPPS) | clustering of z-scored abundance | secondary — attempt if abundance table ships |
| Pathway enrichment (IPA, DAVID, g:Profiler, STRING, NetworKIN) | external web/licensed tools | OUT — external/licensed, not reproducible offline |
Reproduction target (in scope)
Reconstruct the sets the authors fed to venn from Supplementary Tables S1–S28
and recompute the overlap counts, comparing 1:1 to the figure-legend numbers.
This isolates and reproduces the set-overlap (venn) computation — explicitly
NOT the upstream MS quantification/statistics, which are stated out of scope.
CONFIRMED target sheets & numbers (this attempt, 2026-06-22)
The supplementary Excel files literally ship the per-Venn membership tables
(one binary 0/1 indicator column per set, one row per gene=DEpP or
gene_site=DPPS). Region counts = row counts per indicator combination — exactly
what the venn package draws. Reported numbers below read directly off the
figure images (gr1/gr4/gr5.jpg, in original/figures/):
| Figure | Supp sheet (file::sheet) | Reported regions |
|---|---|---|
| Fig 1A DEpP | mmc2 :: Fig1A. AmAc DEpP Venn | only6h 75 / both 89 / only24h 284 (tot 164/373) |
| Fig 1A DPPS | mmc2 :: Fig1A. AmAc DPPS Venn | only6h 108 / both 85 / only24h 424 (tot 193/509) |
| Fig 4A DEpP | mmc12 :: Fig 1A. Human+Mouse DEpP venn | mouse 742 / both 296 / human 228 |
| Fig 4A DPPS | mmc12 :: Fig1A. Human+Mouse DPPS | mouse 2377 / both 142 / human 860 |
| Fig 4E MIC | mmc12 :: Fig1E. MIC vs Human Venn | 914 / 54 / 948 |
| Fig 4E |
Assessments & scoring basis
Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.
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.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
Clean 1:1 reproduction: all 12 in-scope claims (~30 individual Venn region counts across Figs 1A/4A/4E/5A/5F) match the published figures exactly, recomputed deterministically from the authors' own deposited binary membership tables (supp S2-S28). The deviation is zero; nothing is on our side or the authors' side. The upstream MS quantification (Proteome Discoverer + Perseus, licensed GUI), enrichment analyses, and western blots were explicitly out-of-scope and taken as given — a scope limit, not a discrepancy. Note: the BRIEF's GSE171642 is an unrelated ATAC-seq sibling deposit, but the actual phospho data behind every Venn was deposited and used, so data identity is intact.
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-4-8Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.