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

FAM3C/ILEI protein is elevated in psoriatic lesions and triggers psoriasiform hyperproliferation in mice.

EMBO Mol Med · 2023
L1 83/100 3/4
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: Q8 · Severity of the miss (overall human judgment) 🟡
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 +1
✓ What held up
  • Reported values were directly comparable
  • Reported values are derivable from the shared data
  • The central claim held under reproduction
What did not (or only partly)
  • 🟡Could not use the authors’ exact input data
  • 🔴A deviation arose in the data or preprocessing
  • 🔴A deviation was attributed to the published material
  • 🟡The deviation was non-trivial in magnitude
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
83/100
Reproducibility score
0.5 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 61% of all assessed papers rank 430 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 to reproduce the PUBLIC-DATA transcriptomic core, and it reproduces 1:1 with one consistent, cross-validated refinement. This is overwhelmingly a wet-lab/mouse paper (IHC/IF/WB/qPCR/mouse genetics); the RU-listed code (stardist/stardist) is only a Fiji-macro nucleus segmenter applied to non-public stained skin images, so that pipeline has no public input and was NOT attempted (data_unavailable for StarDist). The reproducible computational portion is the DESeq2 re-analysis the Methods describe verbatim on GSE121212. CLEAN RE-RUN (this pass): a freshly-built conda env (R 4.3.3 / DESeq2 1.42.0) and a fresh «our HPC» SLURM job (2218148, COMPLETED) independently re-derived the GSE121212 numbers to 3 significant figures. Sample groups map EXACTLY to the paper (healthy N=38, PSO-lesional=28, PSO-NL=27; EV1D n=66=38+28). RESULT: the paper's central transcriptomic claim that FAM3C/ILEI mRNA is NOT elevated despite elevated protein is REPRODUCED and reinforced; the inducer genes TGFB1/PLAU(uPA)/PLAUR(uPAR) reproduce as significantly elevated (exact match), and canonical psoriasis markers light up massively (20,777 DEGs; pipeline validated). REFINEMENT worth a human's eye: the literal word 'unchanged' is not statistically supported — FAM3C mRNA is slightly but highly-significantly DECREASED (DESeq2 -0.418 log2, padj 6.78e-10; paper's own t-test p=1.55e-8), and in the prior pass this same -0.427 log2 decrease reproduced on a 2nd independent dataset+platform (E-MTAB-8149 microarray, p=5.3e-35). NOT a fabrication concern: the paper presents real violin/box distributions and ran no significance test on the FAM3C panels (legend only tests sibling panels), so 'unchanged' is a visual interpretation of genuine, mildly-decreased data. NOT attempted: StarDist IHC quant (no public images); Fig EV3 GO/GSEA overlap; MINT-PLSDA ILEI-signature (Fig 8/EV5); all wet-lab. No completeness claim. Honest verdict: PARTIAL — the in-scope public-data pipeline reproduces, with a consistent, cross-validated wording refinement on the headline FAM3C-mRNA claim.

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 77
    assessed: 2026-06-16 ⛓ 0a8bd2233744
✎ 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
2026-06-22
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-16
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

ILEI/FAM3C, a cytokine previously implicated only in tumor EMT and metastasis, contributes causally to skin inflammation and drives the psoriasiform phenotype in psoriasis.

Core claims
  • ILEI protein, but not mRNA, is elevated in psoriatic skin and correlates with disease severity finding
  • Inducible keratinocyte-specific ILEI overexpression (K5-ILEIind) causes a psoriasiform phenotype after TPA challenge, with impaired differentiation and neutrophil recruitment finding
  • ILEI activates Erk and Akt signaling, which activates STAT3 via Ser727 phosphorylation mechanism
  • Keratinocyte-specific ILEI deletion ameliorates TPA-induced skin inflammation finding
  • A transcriptomic ILEI signature from K5-ILEIind mice identifies urokinase (uPA) as a targetable regulator of ILEI activity finding
  • Pharmacological uPA inhibition (UK371804) reduces ILEI secretion and improves psoriasiform symptoms in K5-ILEIind mice finding
  • The ILEI gene signature distinguishes psoriasis patients from healthy controls, with uPA among the top separator genes finding
  • K5-ILEIind inducible transgenic mouse model of keratinocyte-specific ILEI overexpression resource
Experimental setups
Assay System Perturbation Readout Platform
Immunohistochemistry (IHC) with chromogenic intensity quantification human skin biopsies (healthy donors and psoriasis patients) none ILEI protein content per keratinocyte, epidermal ILEI load, epidermal thickness
scRNA-seq (published dataset GSE162183) human full-thickness skin (control and psoriasis) none FAM3C mRNA expression across skin cell clusters
Bulk RNA-seq / microarray (published datasets MTAB-8149, GSE121212) human normal and psoriatic (lesional/nonlesional) skin none FAM3C, TGFB1, PLAU, PLAUR, PLG mRNA levels
In vivo TPA/imiquimod inflammatory challenge with histology and IHC K5-ILEIind and ILEIind mouse skin keratinocyte-specific inducible ILEI overexpression (doxycycline diet) + TPA or IMQ epidermal thickness, ILEI protein levels, macroscopic scaly skin phenotype
Immunohistochemistry / immunofluorescence mouse back skin sections ILEI overexpression + TPA Ki67 proliferation marker, Keratin 10 (K10) differentiation marker
Western blot and qPCR mouse total skin extracts and freshly sorted primary keratinocytes ILEI overexpression + TPA Keratin 16 (K16) and Keratin 5 (K5) expression
Ex vivo primary keratinocyte culture with calcium-induced differentiation primary mouse keratinocytes isolated from ILEI-overexpressing and control mice ILEI overexpression + TPA + calcium K10 and loricrin expression (protein and mRNA)
Pharmacological inhibitor treatment TPA-induced K5-ILEIind mice urokinase (uPA) inhibitor UK371804 psoriasiform symptom severity, ILEI secretion
Key results
  • Median ILEI protein content in keratinocytes significantly increased in psoriatic vs healthy skin
  • ILEI intensity in keratinocytes positively correlates with epidermal thickening
  • ILEI mRNA levels unchanged in psoriatic keratinocytes (scRNA-seq) and total skin (bulk RNA-seq)
  • TGFβ, uPA, and uPAR transcripts significantly elevated in psoriasis datasets
  • K5-ILEIind mice show significantly increased epidermal thickening after TPA treatment vs controls
  • K5-ILEIind mice show increased suprabasal Ki67+ proliferating cells
  • ILEI overexpression impairs K10 differentiation marker and increases pathological K16 marker
  • uPA inhibitor UK371804 treatment improves psoriasiform symptoms by reducing ILEI secretion
Key statistics
  • pvalue P < 0.01 (median ILEI protein levels, psoriasis vs healthy keratinocytes (Fig 1C, E))
  • count n = 5,000-15,000 cells per skin sample (ILEI protein content quantification per keratinocyte)
  • count n = 6 healthy, n = 5 psoriasis (human skin IHC cohort)
  • count n = 285 (MTAB-8149 bulk RNA-seq dataset (Fig EV1C))
  • count n = 66 (GSE121212 dataset (Fig EV1D))
  • pvalue P < 0.05, P < 0.001 (epidermal thickness, one-way ANOVA with Tukey test, mouse skin (Fig 2D))
  • pvalue P < 0.0001 (TGFB1, PLAU, PLAUR, PLG expression in normal vs psoriatic lesions, MTAB-8149 (Fig EV1E))
  • correlation Pearson correlation (median ILEI content per keratinocyte vs epidermal thickness in psoriatic skin (Fig 1D))

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 immunohistochemistry (IHC) of human psoriatic and healthy skin with genetically engineered mouse models to investigate ILEI/FAM3C in psoriasis, supplemented by published bulk RNA-seq and scRNA-seq datasets. Two-group comparisons (healthy vs. psoriatic) were made with Student's t-test, multi-group mouse model comparisons used one-way ANOVA with Tukey post-hoc correction, and a Pearson correlation assessed the relationship between ILEI protein levels and epidermal thickness. Results were reported as mean ± SEM with threshold-based asterisk p-value notation throughout.

Replicationmixed Sample sizeHuman IHC: n=6 healthy donors, n=5 psoriatic patients stated explicitly; mouse model: n=2–5 per group across 3 independent experiments stated; bulk RNA-seq public datasets: MTAB-8149 n=285, GSE121212 n=66; mouse time-course experiments n=3 per time point across 2 independent experiments GroupsHealthy vs. psoriatic human skin; K5-ILEI ind vs. ILEI ind controls with doxycycline vs. normal diet and TPA vs. acetone treatment (factorial mouse design); normal vs. nonlesional vs. lesional psoriatic skin in public datasets Pairingunpaired Randomization/blindingnot stated DispersionSEM Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionTukey HSD post-hoc test following one-way ANOVA
Statistical tests used
Test Applied to n Assumptions
Student's t-test (two-group, unpaired) Median ILEI keratinocyte content and total epidermal ILEI load in healthy vs. psoriatic skin (Fig 1C, 1E); mRNA expression of TGFB1, PLAU, PLAUR, PLG in normal vs. psoriatic skin, dataset MTAB-8149 (Fig EV1E) n=6 healthy, n=5 psoriasis (Fig 1C, 1E); n=178 normal skin, n=107 psoriatic lesions (Fig EV1E) not stated
One-way ANOVA with Tukey multiple comparison post-hoc test Epidermal thickness across four experimental groups in K5-ILEI ind mouse model (Fig 2D); mRNA expression of TGFB1, PLAU, PLAUR across normal, nonlesional, and lesional psoriatic skin, dataset GSE121212 (Fig EV1F) n=2–5 per group across 3 independent experiments (Fig 2D); n=38 normal, n=27 nonlesional, n=28 lesional (Fig EV1F) not stated
Pearson correlation Correlation between median ILEI content per keratinocyte and epidermal thickness in psoriatic skin (Fig 1D) n=5 psoriatic patients not stated
Approaches that could also have been used
  • Student's t-test was used to compare ILEI protein levels between healthy (n=6) and psoriatic (n=5) groups
    Could also: Mann-Whitney U (Wilcoxon rank-sum) test — With group sizes of 5–6, normality of the underlying distribution cannot be reliably verified; a non-parametric rank-based alternative makes no distributional assumption and is a common choice at this sample size
  • Pearson correlation was used to relate ILEI content to epidermal thickness across n=5 psoriatic samples
    Could also: Spearman rank correlation — With only 5 data points, bivariate normality and linearity — prerequisites for Pearson — cannot be assessed; Spearman correlation is a non-parametric alternative robust to non-normality and influential outliers at small n
  • Dispersion was reported as SEM throughout
    Could also: Standard deviation (SD) or 95% confidence intervals — SEM decreases with n and primarily conveys precision of the mean estimate rather than spread of individual observations; SD describes biological variability directly, and 95% CIs additionally convey uncertainty about the effect — both are often preferred when sample sizes are small
  • P-values were reported only as threshold-based asterisk categories
    Could also: Exact p-values (e.g., P=0.018) reported alongside or instead of asterisk notation — Exact p-values preserve the full strength-of-evidence information, allow readers to apply alternative thresholds, and facilitate downstream meta-analysis
  • Effect sizes were not reported alongside p-values for the human IHC or mouse model comparisons
    Could also: Cohen's d or standardized mean difference with 95% confidence interval — Effect size metrics quantify the magnitude of a difference independently of sample size, enabling cross-study comparisons and a more complete picture of biological relevance beyond statistical significance alone
  • Multiple independent two-group t-tests were performed across separate figures without a stated overarching multiplicity correction
    Could also: Benjamini-Hochberg FDR correction or Bonferroni correction applied across the set of related two-group comparisons — Applying a correction across a defined family of related tests controls the expected rate of false positives when many comparisons are made within the same study, complementing the within-ANOVA Tukey correction already used
Software: not stated

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
8
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.

GSE121212 GEO in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
GSE162183 GEO in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
HPA050548 HPA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
HPA070796 HPA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet

Downstream reach in the literature

72 downstream papers · 2 datasets

How widely the datasets deposited by this paper are reused across the whole literature (Europe PMC), beyond our assessed set. This is a factual dependency map — reusing a public dataset is normal, good science. It is not a judgement on the downstream papers; the only verdict here is this paper's own, with its cited rationale.

What was reproduced

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

Scope — pmid-37226685

Paper: Malik B, Vokic I, Mohr T, et al. "FAM3C/ILEI protein is elevated in psoriatic lesions and triggers psoriasiform hyperproliferation in mice." EMBO Mol Med 2023. PMID 37226685 · PMCID PMC10331587 · DOI 10.15252/emmm.202216758.

RU-listed code: https://github.com/stardist/stardist (StarDist — a third-party nucleus-segmentation tool). RU-listed data: GEO GSE121212.

What kind of paper is this?

A predominantly wet-lab / mouse-genetics study. The central biology (ILEI/FAM3C protein elevated in psoriasis lesions, psoriasiform hyperproliferation in K5-ILEI mice, TGFβ/uPA axis) rests on IHC/IF, Western blot, qPCR, and mouse phenotyping — none of which is a reproducible computational pipeline on public data. The publicly reproducible computational core is the transcriptomic re-analysis of public datasets.

In scope (pipeline-derived, public data) — ATTEMPTED

The Methods state: "The count matrices of the GSE121212 human dataset … were read into GNU R and analyzed for differentially expressed genes using DESeq2 (Love et al, 2014)." This is a clean, fully-specified pipeline on a public dataset.

id Reported result Paper location Pipeline Public data
EV1D FAM3C/ILEI mRNA unchanged between normal skin (N, n=38) and psoriatic lesions (P-L, n=28) Fig EV1D + text ("ILEI mRNA levels were unchanged … in total skin") DESeq2 on GSE121212 counts GSE121212 (GEO)
EV1F TGFB1, PLAU, PLAUR changed (up in lesional) across N (n=38) / P-NL (n=27) / P-L (n=28) — positive controls of same pipeline Fig EV1F DESeq2 on GSE121212 counts GSE121212 (GEO)
EV3G–I (stretch) GSE121212 DEG volcano + GO enrichment, overlap with mouse "ILEI signature" Fig EV3G–I DESeq2 + GO/GSEA GSE121212

The group counts of GSE121212 confirmed against the series matrix map exactly to the paper: CTRL/healthy=38, PSO-lesional=28, PSO-non-lesional=27 (also AD lesional=21, AD non-lesional=27, AD chronic=6). Fig EV1D n=66 = 38+28 ✓; Fig EV1F = 38/27/28 ✓.

Primary target = EV1D (the headline transcriptomic claim: mRNA unchanged despite elevated protein). EV1F genes serve as positive controls that validate the pipeline (if our DESeq2 reproduces their up-regulation, a non-significant FAM3C is credible, not an artifact of a broken pipeline).

Out of scope (not a reproducible public-data pipeline) — NOT attempted

  • StarDist IHC/DAB quantification (the RU-listed "code"). Used inside a Fiji macro to segment nuclei and measure DAB chromogen intensity on stained human skin sections. The raw images are not deposited anywhere public (only the rendered EMBO figures exist). With no public input images there is nothing to run StarDist on and no value to compare → data_unavailable for this specific pipeline. The tool itself installs/runs fine; it simply has no paper data to reproduce against.
  • E-MTAB-8149 (Fig EV1C/E; microarray, oligo/RMA)DONE (stretch completed): RMA on 402 HuGene-2.1-st CELs reproduced the FAM3C decrease (−0.427 log2, p=5.3e-35) and TGFB1/PLAU/PLAUR elevation — cross-confirming GSE121212. See AUDIT.md.
  • MINT-PLSDA "ILEI signature" ROC/PCA (Fig 8 / EV5): requires the authors' own mouse dataset (E-MTAB-12849) integrated with both human cohorts via mixOmics; complex, multi-step, deferred.
  • All mouse experiments, IHC/IF, WB, qPCR, histology — wet-lab, out of pipeline scope.

Data availability note

The paper's own new data (mouse RNA-seq) is E-MTAB-12849 (ArrayExpress/BioStudies). GSE121212 and E-MTAB-8149 are re-used third-party public datasets. Per BRIEF rule 2 (P16), running an established tool (DESeq2) on the paper's cited public data is a fully valid reproduction.

Figures / tables: Fig EV1DFig EV1CFig EV1F
EV1D-FAM3C
Reported
FAM3C/ILEI mRNA unchanged in psoriatic lesions vs normal skin (GSE121212, Fig EV1D, n=66=38+28)
Reproduced
NOT elevated but a small, robustly significant DECREASE: DESeq2 log2FC=-0.418 (padj 6.78e-10); paper's own Student t-test p=1.55e-8 (Welch 1.07e-8, Wilcoxon 1.62e-8, Cohen d=-1.61); mean log2-norm 10.083(N) vs 9.673(P-L)
partial
EV1F-TGFB1
Reported
TGFB1 significantly elevated in psoriatic lesional skin (GSE121212, Fig EV1F)
Reproduced
ELEVATED: Tukey P-L vs N +0.501 log2 (p.adj 2.05e-6, ANOVA 7.0e-7); DESeq2 log2FC=+0.515 (padj 2.24e-7)
exact
EV1F-PLAU
Reported
PLAU/uPA significantly elevated in psoriatic lesional skin (GSE121212, Fig EV1F)
Reproduced
ELEVATED: Tukey +1.194 log2 (p.adj 4.35e-10); DESeq2 log2FC=+1.195 (padj 1.14e-23)
exact
EV1F-PLAUR
Reported
PLAUR/uPAR significantly elevated in psoriatic lesional skin (GSE121212, Fig EV1F)
Reproduced
ELEVATED: Tukey +1.367 log2 (p.adj 4.35e-10); DESeq2 log2FC=+1.448 (padj 2.66e-30)
exact
pipeline-validity
Reported
(internal positive control) canonical psoriasis-lesional markers up-regulated
Reproduced
DEFB4A +11.38, IL36G +5.02, S100A7 +6.52, KRT16 +5.91, IL17A +6.90, PI3 +8.22, CXCL8 +8.48 (all padj<<1e-40); 20,777 DEGs at padj<0.05 — DESeq2 re-analysis demonstrably faithful
exact
EV1C/E-FAM3C-uPAaxis (E-MTAB-8149, PRIOR PASS, not re-run this pass)
Reported
FAM3C unchanged + TGFB1/PLAU/PLAUR elevated in psoriasis, 2nd dataset E-MTAB-8149 microarray (Fig EV1C/E)
Reproduced
PRIOR-PASS oligo/RMA (archived): FAM3C -0.427 log2, t-test p=5.3e-35 (cross-confirms EV1D to the 2nd decimal on an independent platform); TGFB1 +0.41, PLAU +0.65, PLAUR +0.92 all elevated. Supporting evidence; not freshly re-graded this pass.
partial

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 83/100

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: Q8 · Severity of the miss (overall human judgment) 🟡
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 +1

The publicly reproducible transcriptomic core reproduces 1:1 and even reinforces the paper's central claim: FAM3C/ILEI mRNA is not elevated despite elevated protein, confirmed on two independent cohorts (GSE121212 -0.42 log2; E-MTAB-8149 -0.427), with TGFB1/PLAU/PLAUR elevated and canonical psoriasis markers validating the pipeline. The only substantive deviation is on the authors' interpretation side: the literal word 'unchanged' (FAM3C panels carry no significance test) is contradicted by a small but highly significant decrease — an underspecified-method/significance issue, explicitly not fabrication since the underlying values are fully derivable from shared data. Minor caveats are an E-MTAB sample-subset difference (199/120 vs 178/107) and that the RU-listed StarDist/mouse pipelines have no public input. Overall a solid reproduction with explainable, cross-validated refinements → 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.

464.7 k
tokens (I/O) · 50.9 M incl. cache
164 min
runtime
Per-job HPC accounting not captured for this run — the runtime shown is the reproduction’s measured wall-clock time.