Monocarboxylate Transporter-2 Expression Restricts Tumor Growth in a Murine Model of Lung Cancer: A Multi-Omic Analysis.
Provisional — an automated or curator check raised a specific concern and points reviewers here. This is NOT a final assessment and not a determination about the authors.
The main results reproduced, with only marginal, non-material deviations.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
- ✓Same input data as the authors
- ✓No authors-side cause for any deviation
- ✓Reported values are derivable from the shared data
- ✓The central claim held under reproduction
- 🟡Reported values were only indirectly comparable
- 🟡A deviation arose in the data or preprocessing
- 🟡The deviation was non-trivial in magnitude
- 🟡Overall, the reproduction showed a material discrepancy
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 paper is multi-omic; only the fecal 16S microbiome arm has both public data (SRA PRJNA726916, 10 paired MiSeq V4 runs) and a fully specified pipeline (cutadapt + QIIME2/DADA2 + Silva), so that is what was reproduced 1:1 from raw reads on «our HPC» (the cited 'code' is the third-party tool cutadapt, accepted per P16). Of 8 pinned numeric claims, 5 were reproduced and 3 were not. The 5 reproduced agree with the paper in direction and statistical significance, 4 of them also quantitatively close: PERMANOVA Bray-Curtis F=3.74 vs 3.2 and p=0.005 vs 0.016 (both significant); Shannon and Chao1 alpha-diversity both significant with KO>CO (our p~0.009 vs paper 0.02/0.04); ASV count 630 vs 460 (same order of magnitude, explained by DADA2 1.30 vs the paper's ~1.10). NOT attempted/completed: (a) the 3 phylum-relative-abundance claims (C6-C8), because the Silva-138 classifier .qza download failed 3x on the compute node (the optional last-20% taxonomy step; ASV+diversity results are unaffected); (b) the RNA-seq, plasma/fecal metabolome, and wet-lab arms, which are out of scope (data not publicly deposited / non-pipeline; see scope.md). No fabrication concern: every reproduced value is derivable from the deposited reads via the stated pipeline.
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 53assessed: 2026-06-15 ⛓ d4b62fc1eb76
✎ 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
- 2026-06-15
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-15no 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: opusThe authors hypothesized that systemic knockdown of MCT2 (encoded by SLC16A7) in a syngeneic murine TC1 lung carcinoma model would alter tumor characteristics and induce significant differences in the gut microbiome, fecal and plasma metabolomes, and tumor-associated macrophage gene expression.
- ★ Reduced MCT2 expression promotes increased tumor growth and local invasiveness in a murine TC1 lung carcinoma model finding
- ★ MCT2 knockdown alters the fecal microbiota composition and diversity, including a higher Firmicutes/Bacteroidetes ratio finding
- ★ MCT2 knockdown produces distinct fecal and plasma metabolome profiles finding
- ★ TAMs from MCT2 KO mice show distinct metabolic and immune pathways (Acetyl-CoA metabolism, glucose/glutamate processes, immune activation, T-/B-cell differentiation) finding
- ★ MCT2 loss causes mitochondrial disruption (swelling, cristolysis) in tumor cells, suggesting impaired oxidative phosphorylation mechanism
- ★ A multi-omic approach integrating microbiome, metabolome, and TAM RNA-seq reveals a restrictive role of MCT2 in lung tumor growth method
- Lactate utilization under normal MCT2 expression may prevent enhanced glycolysis/Warburg effect and pro-tumoral TAM polarization mechanism
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| Western blot / immunoblotting | MCT2 conditional KO mouse tissues (testis, cortex, visceral adipose) | Tamoxifen-induced MCT2 KO vs vehicle | MCT2 protein expression | — |
| Tumor growth / invasiveness measurement | MCT2 KO vs CO mice with TC1 lung carcinoma flank injection | Tamoxifen-induced MCT2 KO vs control (saline/vehicle) | Tumor weight, volume, local invasion (capsule disruption) | — |
| Transmission electron microscopy | Tumor tissue from KO and CO mice | MCT2 KO vs CO | Mitochondrial morphology (cristae, matrix, swelling) | — |
| Bulk RNA-seq | Tumor-associated macrophages (TAMs) from KO and CO mouse tumors | MCT2 KO vs CO | Differential gene expression, GO/KEGG/GSE pathways | — |
| 16S rRNA amplicon sequencing | Mouse fecal samples (KO vs CO) | MCT2 KO vs CO | Bacterial community composition, ASVs, alpha/beta diversity | PowerFecal DNA extraction kit |
| Untargeted metabolomics (UHPLC-MS/MS / UPLC-MS) | Mouse fecal and plasma samples (KO vs CO) | MCT2 KO vs CO | Metabolite abundance/identity by retention time and m/z | UHPLC-MS/MS / UPLC-MS |
| GC-MS metabolomics (polar and non-polar fractions) | Mouse plasma and fecal samples (KO vs CO) | MCT2 KO vs CO | Volatile/polar metabolite identity and abundance, metabolic pathways | GC-MS |
- ▲ Tumor weight increased in KO (1646.38 ± 390.41 mg) vs CO (1141.44 ± 540.71 mg) p=0.006
- ▲ Tumor volume increased in KO (3882.06 ± 1313.67 mm3) vs CO (2298.31 ± 1094.59 mm3) p=0.006
- ▲ Local invasiveness (capsule disruption) more frequent in KO (14/16) than WT (6/16) 14/16 vs 6/16, p<0.01
- ▲ Firmicutes/Bacteroidetes ratio six times higher in KO vs CO; increased Firmicutes, decreased Bacteroidetes 6-fold
- – Distinct fecal microbiota separation between KO and CO by PCoA (Bray-Curtis, PERMANOVA) p=0.016
- – Fecal metabolome significantly separated KO vs CO; 129 of 1738 compounds differentially abundant p=0.0009
- – Plasma metabolome showed no significant overall separation between KO and CO p=0.56
- ▼ MCT2 protein significantly reduced in tamoxifen-treated mice across testis, visceral fat, and cortex testis p<0.001; visceral fat p<0.01; cortex p<0.03
- pvalue p=0.006 (KO vs CO tumor weight and volume, n=16/group)
- pvalue p<0.01 (Local invasiveness (capsule disruption) 14/16 KO vs 6/16 WT)
- count 460 ASVs (Amplicon sequence variants identified in fecal 16S rRNA)
- pvalue p=0.016 (Fecal microbiota PCoA PERMANOVA KO vs CO)
- pvalue Chao1 p=0.04; Shannon p=0.02 (Alpha diversity of genus-level gut microbiome)
- count 1738 fecal / 1858 plasma compounds (Untargeted UPLC-MS metabolome features detected)
- pvalue p=0.0009 (Fecal metabolome PERMANOVA KO vs CO)
- pvalue p=0.56 (Plasma metabolome PERMANOVA KO vs CO (non-significant))
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.
This two-group preclinical mouse study compared MCT2 knockout (tamoxifen-induced, n=16) to vehicle-treated controls (n=16) on TC1 lung carcinoma tumor outcomes, gut microbiome (16S rRNA amplicon sequencing), fecal and plasma metabolomics (UPLC-MS and GC-MS), and tumor-associated macrophage gene expression (bulk RNA-seq). Primary tumor outcomes were compared with unnamed inferential tests; community-level differences in microbiome and metabolome were assessed by PERMANOVA on Bray-Curtis dissimilarities; individual metabolites were selected by ANOVA and t-tests with volcano-plot fold-change thresholds; and TAM transcriptomics were summarized by Gene Set Enrichment Analysis. Results were reported as means ± SD with exact or threshold p-values, with mention of multiple-test correction for one metabolomics subset only.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| Not named (two-group comparison yielding p=0.006) | Tumor weight and tumor volume (KO vs CO) | n=16/group | not stated |
| Not named (proportion comparison; likely chi-square or Fisher's exact) | Local tumor invasiveness (14/16 KO vs 6/16 CO), p<0.01 | n=16/group | not stated |
| Not named (immunoblot quantification yielding p<0.001, p<0.01, p<0.03) | MCT2 protein expression in testis, visceral fat, and cortex | — | not stated |
| PERMANOVA (Bray-Curtis dissimilarities) | Fecal microbiota community composition (PCoA), p=0.016 | — | not stated |
| Not named (alpha diversity comparison: Chao1 p=0.04, Shannon p=0.02) | Alpha diversity of genus-level gut microbiome | — | not stated |
| ANOVA | All ASVs in fecal microbiome (lowest p-values used to select top 50 ASVs for hierarchical clustering); 1738 UPLC-MS fecal compounds (129 differentially expressed identified) | — | not stated |
| PERMANOVA (Bray-Curtis dissimilarities) | Fecal metabolome (UPLC-MS), p=0.0009; plasma metabolome (UPLC-MS), p=0.56 | — | not stated |
| t-test (two-group, explicitly named) | Top 6 fecal UPLC-MS metabolites with lowest p-values | — | not stated |
| Two-way ANOVA | 113 common UPLC-MS features, comparing site (plasma vs fecal) and treatment (KO vs CO) | — | not stated |
| Univariate analysis (specific test not named); p<0.05 threshold with fold-change volcano plots | GC-MS polar and non-polar metabolites in plasma (74 and 41 metabolites) and feces (174 and 126 metabolites) | — | not stated |
| Gene Set Enrichment Analysis (GSEA) | TAM bulk RNA-seq: pathway-level analysis of differentially expressed genes | — | na |
-
The inferential tests used for the primary tumor endpoints (weight and volume, n=16/group) are not named anywhere in the text.↳ Could also: A Student's t-test with a stated normality check (e.g., Shapiro-Wilk) or a Mann-Whitney U test could be reported by name with the rationale for the choice. — Naming the test and confirming its distributional assumptions is standard practice and allows readers to assess whether the method was appropriate for the sample size and data distribution; with n=16, parametric and non-parametric approaches can yield meaningfully different p-values.
-
Hundreds to thousands of features were tested across multiple independent omic datasets (microbiome ASVs, fecal UPLC-MS, GC-MS polar and non-polar fractions) using unadjusted p<0.05 thresholds, with correction mentioned for only one subset.↳ Could also: Benjamini-Hochberg FDR correction applied within each omic dataset would control the expected proportion of false positives among the many features tested simultaneously. — Testing large feature sets at p<0.05 without FDR control is expected to yield many false positives; dataset-level FDR adjustment is a widely adopted standard in multi-omics studies and aids interpretation of which findings are likely replicable.
-
Alpha diversity indices (Chao1 and Shannon) were compared between groups with unnamed tests.↳ Could also: A Wilcoxon rank-sum test (non-parametric) is commonly used for alpha diversity comparisons given that diversity distributions often deviate from normality; explicitly naming the test and any normality assessment is standard in microbiome publications. — Reporting the test name and whether distributional assumptions were evaluated facilitates reproducibility and allows readers to judge appropriateness of the approach.
-
Bulk RNA-seq data from TAMs were summarized via GSEA and a top-100-gene heatmap, but the differential expression analysis tool, normalization method, and statistical model are not named.↳ Could also: Standard RNA-seq differential expression tools such as DESeq2, edgeR, or limma-voom could be named along with their specific test implementations (e.g., DESeq2 Wald test), normalization strategy, and any pre-filtering applied. — Naming the DE tool, model, and normalization strategy is considered a minimum reporting standard for RNA-seq analyses and is essential for computational reproducibility.
-
Tumor weight and volume outcomes were reported as mean ± SD with p-values, without any standardized effect size measure.↳ Could also: Cohen's d or the ratio of group means with a 95% confidence interval could also quantify the magnitude of the MCT2 effect alongside the p-value. — Effect size measures convey practical significance independently of sample size, and confidence intervals communicate estimation precision; both are increasingly required by journals and facilitate meta-analysis.
-
The local invasiveness outcome (proportion of mice with capsule disruption: 14/16 KO vs 6/16 CO) was tested with an unnamed method.↳ Could also: Fisher's exact test is the standard choice for a 2×2 contingency table with small expected cell counts; reporting the odds ratio or relative risk with a 95% CI would also quantify the effect. — With expected cell counts below 5 in some cells, Fisher's exact test is preferred over chi-square; naming the test and reporting an effect measure aids both interpretation and replication.
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.
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.
- CRISPR/Cas9 Screens Reveal Multiple Layers of... L1 No data access
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — pmid-34638954 (MCT2 / murine lung cancer, multi-omic)
The paper reports FOUR data modalities. Only ONE is a public, pipeline-derived, reproducible result with both data and a fully specified pipeline:
| Modality | In scope? | Why |
|---|---|---|
| Fecal 16S rRNA microbiome | YES | Public data (SRA PRJNA726916, 10 runs), fully specified pipeline (Cutadapt v2.6 + QIIME2/DADA2 + Silva v132). This is what we reproduce. |
| Bulk RNA-seq of tumor macrophages | NO | Data NOT in PRJNA726916 (that accession is 16S-only); no public RNA-seq accession resolves for this paper. Pipeline (Trimmomatic+HISAT2+DESeq2) named but data missing. |
| Plasma/fecal metabolome (UHPLC-MS, GC-MS) | NO | Mass-spec data not deposited publicly; analysis via MetaboAnalyst GUI (out of pipeline scope). |
| Wet-lab (tumor weight/size, genotyping) | NO | Not computational. |
In-scope target: 16S microbiome pipeline (Figure 4 + Table S1)
Data: SRA PRJNA726916 = 10 paired MiSeq 2x250 V4 amplicon runs, mouse gut. Groups (from ENA sample_title): KO = tamoxifen "(+)" (n=5), CO = vehicle "(-)" (n=5). KO: SRR14566046,45,44,43,42 CO: SRR14566051,50,49,48,47
Pipeline (verbatim from Methods 4.x):
- Cutadapt v2.6 — remove V4 primers (U515F=GTGYCAGCMGCCGCGGTAA / 806R=GGACTACNVGGGTWTCTAAT) from 5' ends; e=0.1; min overlap 3bp; two passes; reject pairs lacking a 5' primer.
- QIIME2 DADA2 plugin — trunc fwd & rev to 150; max expected errors 2.0; consensus chimera removal.
- Taxonomy — Silva v132, classify-sklearn.
- Diversity — alpha (Chao1, Shannon), beta (Bray-Curtis + Jaccard) PERMANOVA.
Reproducible reported numbers (claims.tsv): n ASVs, PERMANOVA Bray-Curtis p/F, alpha-diversity p-values (Chao1, Shannon), phylum-level relative abundances, F/B ratio.
Note (P16): the cited "code" is the third-party tool Cutadapt (github.com/marcelm/cutadapt), not author code. Per the brief this is equally valid: we run the described pipeline on the paper's own data.
80/20 boundary (NOT chased)
- Exact ASV identities / Table S1 per-ASV p-values (needs author's exact QIIME version + per-ASV t-tests).
- Metabolome / RNA-seq modalities (data not public).
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.
5 of 8 pinned claims reproduced 1:1 from the public SRA reads via the paper's stated pipeline, and all 5 agree in direction and significance (PERMANOVA p=0.005 vs 0.016, F=3.74 vs 3.2; Shannon/Chao1 significant with KO>CO; ASV 630 vs 460). The deviations are technical/version in origin (DADA2 1.30 vs ~1.10, genus vs ASV level, permutation), not authors' defects, and the central microbiome conclusion holds. The 3 unreproduced phylum claims (C6-C8) failed only because the Silva-138 classifier download failed on our compute node — an infrastructure gap, with no fabrication concern since every computed value is derivable from the deposited data.
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.