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

Differential Gene Expression and Methylation Analysis of Melanoma in TCGA Database to Further Study the Expression Pattern of KYNU in Melanoma.

J Pers Med · 2022
L1 100/100 PQI 100
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: Q7 · Core claim 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score 0
✓ 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
  • Overall, the reproduction was clean
What did not (or only partly)
  • 🟡The central claim did not (fully) hold under reproduction
How its reproducibility compares
100/100
Reproducibility score
1.5 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 95% of all assessed papers rank 1 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 -> 1:1 EXACT. The paper's 7 DNA-methylation-vs-mRNA Spearman coefficients (KRT18,CDK2,JAK3,BCL2,MITF,MET,KIT) are attributed to GDAC Broad Institute's Correlate_Methylation_vs_mRNA. I obtained that exact pipeline's published output for TCGA-SKCM-TM (Firehose analyses__2016_01_28) on «our HPC»/«infra» and read per-gene Corr_Coeff from the 16049-row matrix. All 7/7 match the reported values to printed precision (paper = Firehose value rounded to 4-5 dp) -> faithful transcription of the cited pipeline, no fabrication signal. P16 third-party-tool reproduction (valid per brief). NOT attempted: independent recompute of Spearman from raw 450k beta + RNA-seq (the optional last-20%; low marginal value after a 7/7 exact match); wet-lab results (WB/qRT-PCR/flow/IHC); Human Protein Atlas; vague DEG claims (KIT '65-fold', '~2-fold' down); GISTIC2 CNV / mutation-rate percentages. The shipped GitHub repo (wmin-debug/...@c21d226) is only a TCGA count-matrix merge utility and emits no comparable numeric claim. Note: paper says 472 SKCM samples but its 3+100+367 breakdown sums to 470 (paper-internal discrepancy).

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 100
    assessed: 2026-06-15 ⛓ 5469ef5e0d2f
✎ 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
2026-06-15
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-15
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

The study aims to analyze and compare melanoma gene expression and methylation profiles in the TCGA database to explore melanoma pathogenesis, and to determine the expression pattern and functional role of KYNU in melanoma, testing whether KYNU can serve as a biomarker and potential therapeutic target.

Core claims
  • KYNU expression is decreased in melanoma despite a high KYNU mutation rate in the TCGA database finding
  • Overexpression of KYNU in low-KYNU melanoma cells promotes tumor cell apoptosis and alters apoptotic (BCL-2), metabolic (KYN, 3-HAA), and invasion/migration (MMP9, E-cadherin) proteins mechanism
  • Key melanoma signaling pathways are EGF/EGFR–RAS–BRAF–MEK–ERK–CyclinD1/CDK4, Ras–PI3K–PTEN–PKB/AKT, and p14/p16(CDKN2A)–MDM2–p53–p21–cyclinD1/CDK4/6–Rb/E2F finding
  • Tumor-promoting genes (MITF, KIT, CDH1, NRAS, AKT1, EGFR, TP53, CDK4) are elevated while PTEN, cAMP, and BCL2 are reduced in melanoma finding
  • Multiple genes (KRT18, CDK2, JAK3, BCL2, MITF, MET, CXCL10, EGF, SOX10, SOCS3, KIT) are negatively regulated by DNA methylation in melanoma finding
  • IL-10 induces immunoregulatory changes and slows the melanoma cell cycle, causing cell-cycle arrest in A375 cells finding
  • Copy number of tumor-promoting genes increased while copy number of tumor suppressor genes decreased in melanoma finding
  • A TCGA-based differential expression/methylation analysis pipeline merging genome-wide multi-sample matrices serves as a resource for studying melanoma target genes resource
Experimental setups
Assay System Perturbation Readout Platform
Differential gene expression analysis (RNA-seq counts) TCGA-SKCM melanoma samples (472 samples: 3 normal, 100 primary, 367 metastatic) none gene expression averages / differential expression between normal and tumor TCGA database / R and Python matrix pipeline
Copy number variation and mutation significance analysis TCGA-SKCM melanoma data none common gene locus mutations and copy number variation GCDA Broad Institute (CopyNumberLowPass_Gistic2)
DNA methylation vs mRNA correlation analysis TCGA-SKCM melanoma data none association strength between methylation and gene expression GCDA Broad Institute (Correlate_Methylation_vs_mRNA)
Immunohistochemistry melanoma and intradermal nevus tissue (also psoriasis, squamous cell carcinoma) none KYNU staining intensity and rate of positive cytoplasmic staining Sigma WH0008942M2 (1:1000) and GeneTex #GTX33291 (1:200) antibodies
Western blotting keratinocytes (HaCaT, HEKα) and melanoma cells (A375, H1205-lu) KYNU overexpression (kynuORF-Pcmv66-Entry plasmid, OriGene #RC214932) protein levels of AKT, ERK1/2, AMPK, p-AMPK, MMP2, MMP9, BCL2 antibodies CST/Santa Cruz/GeneTex; β-actin loading control
qRT-PCR keratinocytes (HaCaT, HEKα) and melanoma cells (H1205-lu, A375) IL-10 (40 ng/mL) and IFN-γ (40 ng/mL) stimulation 24 h mRNA expression of E-cadherin, AKT, ERK1/2 and cell-cycle proteins
CFDA-SE fluorescent staining (proliferation) keratinocytes (HaCaT, HEKα) and melanoma cells (H1205-lu, A375) KYNU overexpression fluorescence intensity reflecting proliferation rate Beyotime CFDA-SE kit C0051; green fluorescence microscope 485 nm
Annexin V–PI / PI single staining flow cytometry melanoma cells (A375, H1205-lu) KYNU overexpression; IL-10 treatment apoptosis and cell-cycle distribution
Key results
  • KYNU level decreased in melanoma while KYNU mutation rate was high in TCGA
  • KYNU overexpression promotes apoptosis in low-KYNU melanoma tumor cells
  • Slower proliferation rate corresponded to stronger CFDA-SE fluorescence intensity
  • MMP9 and AMPK expression decreased in A375, with no obvious change in BCL-2
  • BCL-2 expression decreased significantly in H1205-lu
  • A375 showed cell-cycle arrest indicating IL-10 slows the melanoma cell cycle
  • Tumor-promoting genes (MITF, KIT, NRAS, AKT1, EGFR, TP53, CDK4) elevated; PTEN, cAMP, BCL2 reduced in melanoma
  • Copy number of tumor-promoting genes increased while tumor suppressor gene copy number decreased
Key statistics
  • count 472 samples (3 normal, 100 primary, 367 metastatic) (TCGA-SKCM melanoma transcriptome samples analyzed)
  • count n = 6 untreated and n = 6 vemurafenib-treated (GSE152699 melanoma cell samples)
  • count n = 8 primary, n = 13 repressed BRAF, n = 8 recurrent (GSE152722 mouse tissue samples)
  • other ~15% (KIT mutations in acral and mucosal melanomas)
  • other 10–15% (BRAF or NRAS mutations in acral melanomas)
  • other 0.2% to 8.0% (oral melanoma proportion of all malignant melanomas)
  • other ≈70% (melanomas evolved to vertical growth phase by diagnosis)
  • other 1–8% (melanoma patients who develop multiple primary melanomas)

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 combined bioinformatics analysis of TCGA and GEO publicly available datasets with in vitro cell line experiments to investigate melanoma gene expression, copy number variation, and DNA methylation. TCGA differential expression was screened by comparing Excel-computed mean expression values between normal and tumor samples; GEO dataset KYNU expression differences were assessed with t-tests or one-way ANOVA. Immunohistochemical staining intensity and positivity rates between melanoma and intradermal nevus were compared using an unpaired t-test, and methylation–mRNA associations were derived from the Broad Institute GDAC correlation functions. Results were reported descriptively with qualitative comparisons and some statistical testing, though dispersion measures, exact p-values, and effect sizes were not systematically reported in the available text.

Replicationmixed Sample sizeTCGA: 3 normal, 100 primary melanoma, 367 metastatic; GEO GSE152699: n=6 per group; GEO GSE152722: n=8/13/8 per group; IHC sample size not stated Groupsmelanoma vs. normal tissue (TCGA); KYNU overexpression vs. control in keratinocyte and melanoma cell lines; melanoma vs. intradermal nevus (IHC); treated vs. untreated and recurrence groups (GEO) Pairingunpaired Randomization/blindingnot stated Dispersionnone Effect sizesno Confidence intervalsno Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
unpaired t-test IHC comparison of staining intensity and positive cytoplasmic staining rate between melanoma and intradermal nevus not stated
t-test or one-way ANOVA (choice not further specified) KYNU expression differences in GEO datasets GSE152699 and GSE152722 GSE152699: n=6 per group; GSE152722: n=8 primary, n=13 BRAF-repressed, n=8 recurrent not stated
mean comparison via Excel (no formal inferential test stated) Genome-wide differential gene expression screening between TCGA normal and melanoma samples 3 solid tissue normal, 100 primary melanoma, 367 metastatic melanoma (total n=470 with expression data) na
Correlate_Methylation_vs_mRNA (GDAC Broad Institute function, method not further specified) Association between DNA methylation and mRNA expression for melanoma genes in TCGA SKCM not stated
Approaches that could also have been used
  • TCGA RNA-seq count data differential expression was screened by computing per-group means in Excel and sorting the differences
    Could also: Dedicated RNA-seq differential expression tools such as DESeq2 or edgeR (using negative binomial models on raw counts) or limma-voom could also be applied — These methods model the count-based, overdispersed nature of RNA-seq data, provide shrinkage-estimated fold changes, and yield per-gene statistical significance with FDR-adjusted p-values, which is the widely adopted standard for TCGA count-matrix analyses
  • Multiple genes and multiple group comparisons were evaluated without a stated multiplicity correction
    Could also: Benjamini-Hochberg false discovery rate (FDR) correction or Bonferroni correction could also be applied across the family of tests — When many genes or comparisons are tested simultaneously, a multiplicity correction is commonly used to bound the expected proportion of false positives; FDR control is particularly common in genomics because it is less conservative than Bonferroni while still providing interpretable error-rate guarantees
  • GEO dataset KYNU expression comparisons used a t-test or one-way ANOVA with small group sizes (n=6 or n=8)
    Could also: Non-parametric alternatives such as Mann-Whitney U (for two groups) or Kruskal-Wallis (for three or more groups) with post-hoc Dunn tests could also be used — With small n (6–13 per group), normality assumptions underlying t-tests and ANOVA are difficult to verify; non-parametric rank-based tests make no distributional assumptions and are a common choice in this sample-size range
  • IHC semiquantitative scores (1–4 grade scale) were analyzed with an unpaired t-test
    Could also: Mann-Whitney U test could also be used for ordinal semiquantitative IHC scores — Ordinal scores on a bounded 1–4 scale are not guaranteed to meet the interval-scale and normality assumptions of the t-test; a rank-based test treats the ordinal nature of the data explicitly and is frequently used for this type of IHC scoring
  • Results were reported without dispersion measures (no SD, SEM, or CI stated for the main comparisons)
    Could also: Reporting standard deviation (SD) or 95% confidence intervals alongside group means could also be included — Dispersion measures allow readers to judge variability relative to the effect size and are considered standard for transparent reporting; CI in particular conveys both precision and direction of effect
  • The study used a single unpaired t-test per IHC outcome (staining intensity, positivity rate) to compare two groups
    Could also: A linear mixed model or two-way ANOVA accounting for tissue sample as a random effect (given five randomly selected fields per sample) could also be used — Because five high-power fields were scored per tissue sample, measurements within a sample are not independent; modeling the sample as a random or clustering unit better reflects the nested data structure and can improve inference
Software: R · Python · Microsoft Excel · GDAC Broad Institute (CopyNumberLowPass_Gistic2 / Correlate_Methylation_vs_mRNA) · Human Protein Atlas (online) · NCBI GEO DataSets

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
9
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.

GSE152699 GEO in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
GSE152722 GEO 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-35893303

Paper: Wang et al. 2022, Differential Gene Expression and Methylation Analysis of Melanoma in TCGA Database to Further Study the Expression Pattern of KYNU in Melanoma. J Pers Med 12(8):1209. DOI 10.3390/jpm12081209. PMCID PMC9329910.

What the paper actually did (Methods digest)

A descriptive, multi-tool bioinformatic survey of melanoma (TCGA-SKCM) plus a small wet-lab arm. Datasets named: TCGA-SKCM (472 nevi/melanoma samples: 3 normal, 100 primary, 367 metastatic), GEO GSE152699 (melanoma cells ± vemurafenib, n=6/6), GSE152722 (mouse). Computational results were obtained largely from third-party web tools rather than authors' own code:

  • GDAC Broad Institute Firehose ("GCDA broad institute"): Correlate_Methylation_vs_mRNA (Spearman methylation-β vs mRNA), CopyNumberLowPass_Gistic2, copy-number.
  • KEGG (pathway map05218), Ensembl ID conversion, Human Protein Atlas, GEO DataSets.
  • A python script (the only shipped repo, wmin-debug/TCGA-melanoma-merge-matrix-python-file, commit c21d226) that merges 472 TCGA HTSeq-count files into one matrix (matrix_472.xlsx). It is a data-wrangling utility, not an analysis pipeline; it emits no reported numeric value (no DEG/stat output).

IN SCOPE (pipeline-derived, attempted)

result reported (loc) pipeline how reproduced
Methylation–mRNA Spearman corr for 7 genes KRT18 −0.8219, CDK2 −0.71693, JAK3 −0.66835, BCL2 −0.65555, MITF −0.65016, MET −0.59344, KIT −0.5003 (Results §3.3 / methylation section) GDAC Firehose Correlate_Methylation_vs_mRNA, SKCM-TM obtain the exact named-pipeline Level_4 output (analyses__2016_01_28, SKCM-TM) on the paper's data and read corr_Coeff per gene

This is the clearest, most precise claim set (4–5 significant figures) and names its pipeline exactly → the highest-value, lowest-ambiguity reproduction target. Per BRIEF rule 2 (P16), applying the named third-party tool to the paper's data is a fully valid reproduction.

OUT OF SCOPE (not attempted, with reason)

  • Wet-lab: Western blot, qRT-PCR, flow cytometry (apoptosis/proliferation), IHC (KYNU in melanoma n=8 vs nevi n=9), cell-line A375/H1205-lu — non-computational.
  • Web-tool point lookups w/o reproducible pipeline spec: "KIT 65-fold higher", "PTEN/cAMP/BCL2 ~2-fold down" (vague, no thresholds/file), Human Protein Atlas images, KEGG pathway membership — no_expected_result/non_pipeline for repro.
  • CNV / mutation rates (BRAF 54/452 up etc., GISTIC2 arm/focal): derivable from Firehose but reported as descriptive percentages tied to the metastatic cohort selection; lower precision, deprioritized as the optional last-20% (BRIEF rule 3).
  • Repo merge script: could be re-run on the 472 GDC count files, but produces no comparable reported number → excluded as a claim (noted, not graded).

Decision

Reproduce the 7 methylation–mRNA Spearman coefficients via the exact Firehose pipeline output the paper cites. One «our HPC» SLURM job (download+extract+lookup on «infra»). Everything else: documented as out-of-scope above.

methcorr_KRT18
Reported
-0.8219
Reproduced
-0.8219036
exact
methcorr_CDK2
Reported
-0.71693
Reproduced
-0.7169284
exact
methcorr_JAK3
Reported
-0.66835
Reproduced
-0.6683464
exact
methcorr_BCL2
Reported
-0.65555
Reproduced
-0.6555525
exact
methcorr_MITF
Reported
-0.65016
Reproduced
-0.6501631
exact
methcorr_MET
Reported
-0.59344
Reproduced
-0.5934391
exact
methcorr_KIT
Reported
-0.5003
Reproduced
-0.5003042
exact

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 100/100

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 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score 0

All 7 methylation-vs-mRNA Spearman coefficients reproduce exactly to printed precision by reading the paper's explicitly cited GDAC Firehose Correlate_Methylation_vs_mRNA SKCM-TM output; the sole deviation is display rounding (e.g. -0.8219036 → -0.8219), so derivability and severity are clean on the authors' side. This is a faithful third-party-tool transcription with no fabrication signal. The caveat is scope: the paper's central KYNU/DEG/wet-lab conclusions were out of scope and untested, so q7 is limited rather than fully confirmed. A minor paper-internal sample-count discrepancy (472 vs 470) is noted but does not affect the reproduced numbers.

🤝
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.

95.1 k
tokens (I/O) · 5.6 M incl. cache
20 min
runtime · 0 CPU-h
0 GB
peak RAM
1
HPC jobs
hummel
machine