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

The manganese transporter SLC39A8 links alkaline ceramidase 1 to inflammatory bowel disease.

Nat Commun · 2024
L1 70/100 3/4
⚑ Flagged for review — a reproduced result did not match the reported value

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.

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: Q5 · Derivability / plausibility 🟡
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 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +4
✓ What held up
  • Same input data as the authors
  • Reported values were directly comparable
  • The central claim held under reproduction
What did not (or only partly)
  • 🟡A deviation arose in the data or preprocessing
  • 🔴A deviation was attributed to the published material
  • 🟡Reported values were not (fully) derivable from the shared data
  • 🟡The deviation was non-trivial in magnitude
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
70/100
Reproducibility score
at the mean
vs. all fields · 1173 studies
🎯 Scores higher than 37% of all assessed papers rank 732 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 for the RNA-seq headline; PARTIAL overall. The repo (SeoResearchLab/IECKO2023) ships DATA ONLY (one lipidomics .xlsx) and NO analysis code, so reproduction = standard tools on the paper's own data (P16). C1 RNA-seq: GEO GSE192695 ships the authors' processed expected-count matrix; DESeq2 (~tissue+genotype, ISKO vs WT, padj<0.1) reproduces the paper's central transcriptomic claim 1:1 -- EXACTLY 4 DEGs: Slc39a8 down + Acer1/Ighv1-55/Entpd4b up, matching 'four genes, only Slc39a8 down, Acer1 among up-regulated'. C2: the shipped ceramide_SPB t-test numbers are exactly re-derivable from the shipped raw matrix (group-mean corr 1.000, p corr 0.998) -> shipped stats are genuine (anti-fabrication check passes). C3 (Fig 8k '40 lipids P_adj<0.2' and 'SMs all downregulated'): NOT reproduced -- a standard t-test+BH on the full tissue-wt-normalized panel yields 204 (FDR<0.2) / 47 (raw p<0.05), and SMs are 28/56 down, not all; the authors' exact full-panel filtering is undocumented because no code was deposited. NOT attempted (out of scope/80-20): STAR alignment from raw FASTQ (counts already deposited), MS peak-picking/lipid ID (proprietary upstream), and all wet-lab phenotyping. No fabrication detected; the Fig 8k gap is under-specification, not fabricated values.

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 70
    assessed: 2026-06-15 ⛓ 2c51d131d925
✎ 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 authors hypothesized that intestinal epithelial SLC39A8 controls intestinal Mn homeostasis and epithelial integrity, and that dysregulation of Mn homeostasis disrupts the epithelial barrier, thereby predisposing to inflammatory bowel disease (IBD).

Core claims
  • Intestinal epithelial cell-specific deletion of Slc39a8 causes systemic manganese (Mn) deficiency in blood and multiple organs. finding
  • Intestinal epithelial SLC39A8 is essential for in vivo intestinal absorption of dietary Mn. finding
  • SLC39A8 localizes to the apical membrane of enterocytes and mediates 54Mn uptake from the luminal/apical side. mechanism
  • Unbiased RNA-seq identifies alkaline ceramidase 1 (ACER1) as a target affecting intestinal epithelial integrity in Slc39a8-IEC KO intestine. finding
  • Treatment with an ACER1 inhibitor attenuates colitis in Slc39a8-IEC KO mice by remedying barrier dysfunction and enhancing tight junction proteins. finding
  • IEC-specific deletion of Slc39a8 exacerbates DSS-induced colitis following intestinal epithelial injury. finding
  • Generation of Slc39a8-IEC KO mice (Slc39a8 fl/fl x Villin-Cre) and intestinal organoid monolayer cultures as models to study Mn transport and epithelial integrity. resource
  • ACER1 inhibition is proposed as a therapeutic target for IBD associated with impaired Mn homeostasis/SLC39A8 deficiency. mechanism
Experimental setups
Assay System Perturbation Readout Platform
qPCR (gene expression) C57BL/6J mouse tissues and control vs Slc39a8-IEC KO mouse intestinal/other tissues; Slc39a8-IEC KO-derived enteroids Slc39a8 IEC-specific KO (Villin-Cre) Slc39a8 and metal transporter (Slc11a2/DMT1, Slc39a14/ZIP14, Slc30a10/ZnT10, Slc40a1/FPN) mRNA levels
Immunofluorescence/confocal microscopy Frozen intestinal sections (duodenum, jejunum, ileum, colon) of C57BL/6J mice; enteroid/colonoid monolayers none / Slc39a8 KO SLC39A8, ZO-1 protein localization (apical membrane)
ICP-MS metal quantification Tissues (intestine, liver, bile, lung, kidney, heart, brain, whole blood) from control and Slc39a8-IEC KO mice Slc39a8 IEC-specific KO Mn, zinc, iron, copper, selenium concentrations ICP-MS
Radiotracer 54Mn absorption (oral-intragastric gavage) Control and Slc39a8-IEC KO mice (10 weeks) Slc39a8 IEC-specific KO; 54Mn gavage 54Mn radioactivity (cpm) in blood, enterocytes, liver, lung, kidney, heart, femur γ-counter
Radiotracer 54Mn clearance (intravenous tail vein injection) Control and Slc39a8-IEC KO mice Slc39a8 KO; IV 54Mn 54Mn radioactivity (cpm) in blood and tissues γ-counter
54Mn uptake assay in organoid monolayers (apical and basolateral) Intestinal organoid (enteroid/colonoid) monolayer cultures from control and Slc39a8-IEC KO mice Slc39a8 KO; 54Mn added to apical or basolateral chamber cell-associated 54Mn radioactivity gamma counter
Unbiased transcriptomic analysis (RNA-seq) Slc39a8-IEC KO intestine Slc39a8 IEC-specific KO differential gene expression (identified ACER1)
DSS-induced experimental colitis Control and Slc39a8-IEC KO mice Dextran sodium sulfate (DSS); ACER1 inhibitor treatment colitis severity / intestinal injury, barrier permeability, tight junction proteins
Key results
  • Slc39a8 mRNA markedly reduced in Slc39a8-IEC KO intestine: duodenum, jejunum, ileum, colon -90% duodenum, -90% jejunum, -93% ileum, -98% colon
  • Whole blood and tissue Mn (ileum, colon, liver, lung, heart, brain) substantially reduced in Slc39a8-IEC KO mice; zinc, iron, copper, selenium unchanged
  • 54Mn in whole blood lower in KO mice 15 min after oral gavage -33% (P<0.05)
  • 54Mn in ileum reduced in KO mice after oral gavage -44% (P<0.05)
  • 54Mn reduced in liver, lung, kidney, heart, femur after oral gavage in KO mice liver -61%, lung -59%, kidney -81%, heart -69%, femur -41% (all P<0.05)
  • After IV 54Mn injection, blood and tissue 54Mn did not differ between KO and control (excludes increased clearance)
  • Apical 54Mn accumulation impaired in Slc39a8-IEC KO enteroid and colonoid monolayers; basolateral uptake unchanged enteroid -35% (P<0.01), colonoid -41% (P<0.001)
  • Slc39a8-IEC KO enteroids show increased Slc11a2/DMT1 and decreased Slc39a14/ZIP14, Slc30a10/ZnT10, Slc40a1/FPN transcripts
Key statistics
  • fold_change -98% (P<0.001) (Slc39a8 mRNA reduction in colon of Slc39a8-IEC KO mice)
  • fold_change -44% (P<0.05) (54Mn reduction in ileum after oral gavage (control n=4, KO n=4))
  • fold_change -33% (P<0.05) (54Mn reduction in whole blood 15 min after oral gavage)
  • fold_change -81% (P<0.05) (54Mn reduction in kidney after oral gavage in KO mice)
  • fold_change -35% (P<0.01) (Apical 54Mn accumulation reduction in KO-derived enteroid monolayers)
  • fold_change -41% (P<0.001) (Apical 54Mn accumulation reduction in KO-derived colonoid monolayers)
  • count n=6 per group (C57BL/6J mice per group for qPCR tissue expression of Slc39a8)
  • count n=5 male, n=6 female per group (ICP-MS Mn analysis in control and Slc39a8-IEC KO mice at 20 weeks)

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 study used a conditional knockout mouse model (Slc39a8-IEC KO) and intestinal organoid monolayer cultures to investigate the role of intestinal epithelial SLC39A8 in manganese homeostasis and epithelial integrity. Primary comparisons between knockout and control animals were made with unpaired two-tailed Student's t-tests for most continuous outcomes (metal concentrations by ICP-MS, radiotracer counts, organoid uptake) and one-way ANOVA with Bonferroni post-hoc correction for multi-tissue gene expression data. Results throughout are reported as mean ± SEM with p-values expressed as threshold inequalities.

Replicationbiological Sample sizePer-figure n values stated in legends; no formal power calculation or justification described in the visible text GroupsSlc39a8-IEC KO mice vs. floxed control mice; male and female cohorts reported separately for some comparisons Pairingunpaired Randomization/blindingnot stated DispersionSEM Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionBonferroni's multiple comparisons test (post-hoc to one-way ANOVA) for Fig. 1c; no correction stated for the multiple unpaired t-tests across figures
Statistical tests used
Test Applied to n Assumptions
One-way ANOVA with Bonferroni's multiple comparisons post-hoc test Fig. 1c — qPCR of Slc39a8 mRNA across multiple intestinal segments and non-intestinal tissues in control vs. Slc39a8-IEC KO mice n = 4 per group not stated
Unpaired two-tailed Student's t-test Fig. 1d — ICP-MS metal concentrations (Mn, and by implication Zn, Fe, Cu, Se in Supplementary Figs. 2a–d) across tissues in control vs. Slc39a8-IEC KO mice, stratified by sex n = 5 male per group; n = 6 female per group not stated
Unpaired two-tailed Student's t-test Fig. 2a–f — radiotracer 54Mn counts in blood, intestinal segments, and peripheral tissues after oral gavage or intravenous injection in control vs. Slc39a8-IEC KO mice n = 4 per group (n = 2 male, n = 2 female each) not stated
Unpaired two-tailed Student's t-test Fig. 3d — apical 54Mn accumulation in enteroid and colonoid monolayer cultures from control vs. Slc39a8-IEC KO mice n = 3 biologically independent samples not stated
Unpaired two-tailed Student's t-test Fig. 3f — basolateral 54Mn accumulation in enteroid and colonoid monolayer cultures from control vs. Slc39a8-IEC KO mice n = 3 biologically independent samples not stated
Approaches that could also have been used
  • Dispersion is reported throughout as mean ± SEM with small group sizes (n = 3–6)
    Could also: Mean ± SD, or mean with 95% confidence intervals, could also convey the spread of the data — With small n, SD gives readers a direct sense of between-animal variability; 95% CIs additionally communicate the uncertainty around the estimated mean and are increasingly favoured in reporting guidelines for preclinical studies
  • Multiple independent unpaired t-tests were used across many tissue/organ comparisons within the same experiment (e.g., Figs. 1d, 2b–c, 2e–f)
    Could also: A one-way or two-way ANOVA (with tissue/organ as a factor) followed by a post-hoc correction (e.g., Tukey HSD or Holm-Bonferroni) could also be applied to the full family of comparisons — Treating the organ-level comparisons within a single experiment as a family and correcting jointly would formally control the experiment-wise error rate across those simultaneous tests
  • Unpaired t-tests were used for groups of n = 3–4 per group (Figs. 2a–f, 3d, 3f) without stated normality assessment
    Could also: A non-parametric alternative such as the Mann-Whitney U test (Wilcoxon rank-sum) could also be applied when sample sizes are too small to assess normality reliably — With n as low as 3, the central-limit-theorem rationale for t-test robustness is limited; a rank-based test makes no distributional assumption, which is an option some journals require for very small n
  • Sex was treated as a stratification variable (separate male/female panels) rather than a formal factor in the analysis of metal concentrations (Fig. 1d)
    Could also: A two-way ANOVA with genotype and sex as factors could also model the sex × genotype interaction formally — Including sex as a crossed factor tests whether the knockout effect differs by sex and can increase statistical power by accounting for sex-related variance; NIH guidelines increasingly encourage this approach in preclinical studies
  • P-values are reported only as threshold inequalities (< 0.001, < 0.05, < 0.01)
    Could also: Exact p-values (e.g., p = 0.023) could also be reported — Exact p-values allow readers and meta-analysts to apply different alpha thresholds, perform power calculations, and aggregate findings across studies; reporting guidelines such as APA and many journals now recommend exact values
  • The unbiased RNA-seq transcriptomic analysis is mentioned in the abstract and methods narrative but statistical details (differential expression method, FDR threshold, fold-change cutoff) are not present in the available text excerpt
    Could also: Standard workflows such as DESeq2 (negative-binomial Wald test with Benjamini-Hochberg FDR) or edgeR could be applied; the chosen method and thresholds would ideally be stated explicitly — Full disclosure of the DE pipeline, normalisation strategy, and multiple-testing correction threshold allows reproducibility assessment and comparison with other transcriptomic datasets
Software: Not stated in visible text

Result convergence & founder nodes

Findings this paper shares with others that ran a comparable experiment. A node’s strength is how many independent papers report it (replication breadth) — not how often it is cited, so a heavily-replicated but under-cited founder still stands out.

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
28
Impact: medium
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.

What was reproduced

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

Scope — pmid-38839750

Paper: Choi et al. 2024, Nat Commun 15:4775. "The manganese transporter SLC39A8 links alkaline ceramidase 1 to inflammatory bowel disease." DOI 10.1038/s41467-024-49049-8 · PMCID PMC11153611.

Model: intestinal-epithelial-cell-specific Slc39a8 knockout (Slc39a8-IEC-KO, Slc39a8-fl/fl; Villin-Cre) vs control (Slc39a8-fl/fl).

Artifacts available

  • Code repo github.com/SeoResearchLab/IECKO2023 (commit 765bc5c, default branch main, public, not archived). Contains NO analysis code — a single file EX01243_LipidomicsData.xlsx (the lipidomics dataset). So there is no authors' pipeline script to run; reproduction = applying standard tools to the shipped data per the Methods text (brief rule P16: valid).
  • Data GEO GSE192695 — RNA-seq of ileum+colon, control vs Slc39a8-IEC-KO, n=3/group, 12 samples. Ships a processed expected-count matrix (GSE192695_gene_expected_count.txt.gz, RSEM/STAR, Ensembl GRCm38 gene IDs).

In scope (pipeline-derived, attempted)

id result pipeline location
C1 "four genes with altered expression (P_adj<0.1); only Slc39a8 down, Acer1 among up-regulated" STAR→RSEM counts → DESeq2 (Methods). Counts are deposited → DESeq2 only. Fig 6a; Results "Transcriptome analyses…"; GEO summary
C2 ceramide/sphingoid-base statistics (group means, t-test p, FDR) for 102 species t-test on ln-transformed intensities (authors ship the t.test + ceramide_SPB sheets) Excel t.test sheet (internal-consistency / anti-fabrication check)
C3 "40 differentially regulated lipids (P_adj<0.2)"; "sphingomyelins were all significantly downregulated" t-test + FDR on full tissue-wt-normalized lipidome Fig 8k; Results lipidomics paragraph

Out of scope (not attempted, why)

  • All wet-lab / phenotyping: histology, IBD/colitis scoring, Mn measurements (ICP-MS), ceramidase enzyme assays, organoid imaging, body-weight/survival — manual/bench, no pipeline.
  • STAR alignment from raw FASTQ — unnecessary; GEO ships the expected-count matrix the authors used; running STAR would re-derive identical inputs at large compute cost (80/20).
  • Lipid identification (LIPIDBLAST/MultiQuant peak picking) — upstream proprietary MS processing; shipped Excel already contains the processed intensity matrices.

Compute

All on «our HPC» («infra» «our HPC»-2 SLURM, kubisch_std). «infra» work dir: «path». No data on «host» (only small result CSVs).

Figures / tables: Fig 6aFig 8k
C1a
Reported
4 DEGs in Slc39a8-IEC-KO (P_adj<0.1)
Reproduced
4 DEGs
exact
C1b
Reported
Slc39a8 downregulated
Reproduced
log2FC -1.13, padj 1.25e-5 (down)
exact
C1c
Reported
Acer1 upregulated
Reproduced
log2FC +0.84, padj 0.046 (up)
exact
C1d
Reported
4 genes (Slc39a8 down + Acer1 up named)
Reproduced
Slc39a8 down; Acer1, Ighv1-55, Entpd4b up
within tolerance
C2
Reported
shipped ceramide t-test values (authenticity)
Reproduced
group-mean corr=1.000 (diff 0.0); p corr=0.998 (max diff 0.064)
within tolerance
C3a
Reported
40 differentially regulated lipids (Fig 8k, P_adj<0.2)
Reproduced
204 full-panel / 139 identified-only at FDR<0.2; 47 at raw p<0.05
did not match
C3b
Reported
sphingomyelins all significantly downregulated
Reproduced
28/56 SM down; 5-11 significant
did not match

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 70/100

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: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +4

The paper's central RNA-seq claim reproduces 1:1 from deposited GEO counts (exactly 4 DEGs: Slc39a8 down, Acer1/Ighv1-55/Entpd4b up), and an anti-fabrication check confirms the shipped ceramide t-test values are exactly re-derivable (mean corr 1.000, p corr 0.998). The one substantive deviation is Fig 8k ('40 lipids P_adj<0.2' and 'SMs all downregulated'), which does not reproduce (204/47 lipids; 28/56 SM down) — this sits on the authors'/data-availability side because the repo is a data-only deposit with no analysis code, leaving the exact filtering under-specified rather than fabricated. Overall a solid reproduction of the core conclusion with an explainable, secondary lipidomics discrepancy → yellow.

🤝
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.

134 k
tokens (I/O) · 7.7 M incl. cache
15 min
runtime · 0.01 CPU-h
1 GB
peak RAM
1
HPC jobs
hummel
machine