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

Transcriptomic Adjustments of Staphylococcus aureus COL (MRSA) Forming Biofilms Under Acidic and Alkaline Conditions.

Front Microbiol · 2019
L1 73/100 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: Q7 · Core claim 🟡
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 +5
✓ What held up
  • Same input data as the authors
  • Reported values were directly comparable
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
  • 🟡The central claim did not (fully) hold under reproduction
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
73/100
Reproducibility score
at the mean
vs. all fields · 1173 studies
🎯 Scores higher than 41% of all assessed papers rank 664 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 DOWNSTREAM analysis 1:1. The 'code' link is the third-party TM4/MeV tool (P16), not author code; GEO ships the authors' own normalised value matrix (GSE138075 series matrix, 3887 probesets x 12 arrays) which is exactly the input to their Excel step. Replaying the described computation (log2 of mean expression ratio + two-tailed paired t-test, p<0.05) on that matrix reproduces EVERY reported per-gene number: all 92 log2FC values in Tables 1A-4B match a probeset to <=0.05 (most <=0.01) and 84/92 also match the printed p to <=0.001; the named genes (codY, mecA, ctsR, sceD, femA, agrB, sarA, hfq) match their annotated probeset on BOTH log2FC and p exactly. => the reported values are genuine, NO fabrication. What does NOT reproduce is the DEG COUNTS: the stated thresholds (log2FC>~3.1 & p<0.05) actually pass 143-196 probesets per comparison (e.g. 177 up / 167 down for pH9 biofilm, only 2 AFFX controls), not the reported 8/11/16/16 up or 4/18/11/12 down. The reported short lists are a heavily curated, annotation-driven subset (reported genes have variance ranks 191-564, not top-50; significant probesets up to log2FC 23.6 go unreported; even band+presence filtering leaves 154 candidates). The 'MeV variance filter value 50' + manual KEGG/Aureowiki curation that shrinks ~175 to ~8-18 is under-specified and not algorithmically reproducible -- selective reporting, not fabrication. NOT attempted: independent RMA re-normalisation from raw CEL (blocked -- no Bioconductor CDF for GPL1339, would need Thermo S_aureus.CDF + makecdfenv; low value since deposited values already reproduce the numbers), and all wet-lab results (growth/CFU/biofilm assays/RT-PCR/microscopy, out of scope).

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 73
    assessed: 2026-06-16 ⛓ 917f77eda1c2
✎ 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
2026-06-16
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-18
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

To detect genes differentially expressed in Staphylococcus aureus COL (MRSA) biofilm-associated and planktonic cells under acidic (pH5) and alkaline (pH9) conditions using DNA microarrays, in order to understand the molecular mechanisms linking pH-related stress response with biofilm formation and pathogenicity.

Core claims
  • S. aureus COL (MRSA) can survive and grow under acidic (pH5) and alkaline (pH9) conditions both planktonically and as a biofilm. finding
  • The pathogen is possibly more tolerant to highly alkaline than acidic environments. finding
  • Genes encoding transcription regulators, ion transporters, cell wall biosynthetic enzymes, autolytic enzymes, adhesion proteins and antibiotic resistance factors are differentially regulated by pH and growth mode, most associated with biofilm formation. finding
  • Microarray-based transcriptomic profiling of planktonic and biofilm cells at acidic and alkaline pH identifies pH/growth-mode-specific gene expression adjustments. method
  • Eight genes were over-expressed in biofilm cells at alkaline pH9, including transcriptional regulators CodY, MecA, CtsR and capsule enzyme CapC. finding
  • Eleven genes were over-expressed in biofilm cells at acidic pH5, including cell surface proteins MapW, Efb/FnbA and secreted VWbp. finding
  • FemA (factor essential for methicillin resistance) was over-expressed in planktonic cells at both pH9 and pH5. finding
  • These results facilitate development of new treatment or disinfection strategies against biofilm-associated MRSA. resource
Experimental setups
Assay System Perturbation Readout Platform
Growth/viability assay (CFU enumeration by serial dilution) S. aureus COL (MRSA) planktonic cells in liquid TSB acidic/neutral/alkaline pH (pH5, 7, 9) adjusted with HCl/NaOH colony forming units per 10 mL
Growth/viability assay (CFU enumeration by serial dilution) S. aureus COL (MRSA) biofilm on nitrocellulose membrane on solid TSA acidic/neutral/alkaline pH (pH5, 7, 9) colony forming units per nitrocellulose disk nitrocellulose membrane 0.45 μm (Sartorius)
DNA microarray (transcriptomics) S. aureus COL planktonic cells in liquid TSB acidic (pH5) and alkaline (pH9) vs neutral (pH7) gene expression (Log2 fold change) GeneChip S. aureus Genome Array (Affymetrix Cat. No. 900514)
DNA microarray (transcriptomics) S. aureus COL biofilm cells on nitrocellulose on solid TSA acidic (pH5) and alkaline (pH9) vs neutral (pH7) gene expression (Log2 fold change) GeneChip S. aureus Genome Array (Affymetrix Cat. No. 900514)
Total RNA extraction and first-strand cDNA synthesis S. aureus COL biomass (planktonic and biofilm) none RNA quality, cDNA for hybridization Nucleospin RNA II kit (Macherey-Nagel); PrimeScript 1st strand cDNA Synthesis Kit (Takara)
Key results
  • Total planktonic growth reached 10^9 CFU/10 mL in acidic and 10^10 CFU/10 mL in neutral and alkaline media, indicating greater tolerance to alkaline conditions 10^9 vs 10^10 CFU/10 mL
  • Eight genes over-expressed in biofilm cells at pH9 log2 fold-change > 3.16
  • Eleven genes over-expressed in biofilm cells at pH5 log2 fold-change > 3.12
  • Sixteen genes over-expressed in planktonic cells at pH9 log2 fold-change 3.22–4.14
  • Sixteen genes over-expressed in planktonic cells at pH5 log2 fold-change 2.74–4.12
  • Four genes down-regulated in biofilm cells at pH9 (AirR, NreB, PhoU-related, MapW) log2 fold-change -3.24 to -3.34
  • Eighteen genes down-regulated in biofilm cells at pH5 log2 fold-change -2.37 to -4.05
  • FemA over-expressed in planktonic cells at both pH9 and pH5 4.12 log2 fold-change
Key statistics
  • pvalue 0.0009 (p-value for total planktonic/biofilm growth difference across pH)
  • fold_change log2 > 3.16, p < 0.0061 (8 upregulated genes in biofilm cells at pH9)
  • fold_change log2 > 3.12, p < 0.0316 (11 upregulated genes in biofilm cells at pH5)
  • fold_change log2 3.22–4.14, p < 0.0356 (16 upregulated genes in planktonic cells at pH9)
  • fold_change log2 2.74–4.12, p < 0.0354 (16 upregulated genes in planktonic cells at pH5)
  • fold_change log2 -3.24 to -3.34, p < 0.0042 (down-regulated genes in biofilm cells at pH9)
  • count over 3,300 ORFs (open reading frame genes on the microarray)
  • count 72,444 (MRSA infection cases reported in US in 2014 (morbidity 11.8%))

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 used Affymetrix GeneChip microarrays to profile genome-wide transcriptional changes in S. aureus COL cells growing as biofilm or planktonically at pH 5, 7, and 9, with biological duplicates (n = 2) for microarray experiments and at least three biological replicates for colony-forming unit (CFU) counts. Raw microarray data were normalized in R (TM4 protocol), variance-filtered in MeV, and differential expression was assessed by computing Log2 fold-change ratios of averaged expression values and applying a two-tailed paired t-test in Excel. Genes were reported as significantly differentially expressed when Log2 fold-change exceeded approximately 3.1 and p < 0.05; no correction for multiple testing across the >3,300 array features was stated.

Replicationbiological Sample sizeMicroarray: n = 2 biological duplicates per condition; CFU counts: at least n = 3 biological replicates per condition; no power calculation stated GroupsBiofilm and planktonic S. aureus COL cells at pH 5, 7, and 9 (six conditions); pairwise comparisons to pH 7 neutral control Pairingpaired Randomization/blindingnot stated DispersionSD Exact p-valuesyes Confidence intervalsno Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
Two-tailed paired t-test Microarray gene expression comparisons between pH conditions (pH9 vs pH7; pH5 vs pH7) in both biofilm and planktonic cells n = 2 biological replicates per condition not stated
Unspecified test (p-value = 0.0009 reported) Comparison of total growth (log CFU) across pH conditions in liquid and solid media (Figure 1) n = 3 biological replicates not stated
Approaches that could also have been used
  • Differential expression across >3,300 microarray features was assessed with a paired t-test at p < 0.05 with no correction for multiple comparisons.
    Could also: Apply a false discovery rate procedure (e.g., Benjamini-Hochberg FDR) across all tested probes simultaneously, as implemented in limma (R) or similar microarray-specific tools. — When thousands of features are tested simultaneously, the expected number of false positives at an uncorrected α = 0.05 can be large; FDR control provides a principled way to interpret the resulting gene list and is the current standard in transcriptomics
  • Microarray experiments used n = 2 biological replicates per condition.
    Could also: Use three or more biological replicates per condition, which is the commonly cited minimum for microarray and RNA-seq studies. — Additional replicates improve variance estimation and statistical power; with n = 2, degrees of freedom for the t-test are minimal, limiting the ability to detect true differences and increasing sensitivity to outliers
  • Differential expression was assessed using a standard paired t-test computed in Excel.
    Could also: Use a purpose-built microarray analysis package such as limma (R), which implements moderated t-statistics (empirical Bayes shrinkage of variance estimates across genes). — Moderated statistics borrow information across the full gene set to stabilize per-gene variance estimates, which is particularly beneficial when n is small; this approach is widely used and better calibrated for microarray data than the standard t-test
  • The statistical test used to compare CFU counts across pH conditions (Figure 1, p = 0.0009) was not identified in the text.
    Could also: Name the specific test (e.g., one-way ANOVA followed by Tukey HSD, or Kruskal-Wallis followed by Dunn's test) and report the test statistic alongside the p-value. — Identifying the test and reporting the statistic (e.g., F or H value) allows readers to assess whether model assumptions were met and enables independent reproduction of the analysis; for three-group comparisons, an omnibus test with post-hoc correction also accounts for multiple pairwise contrasts
  • Results were reported with exact p-values and Log2 fold-change ratios but without confidence intervals.
    Could also: Report 95% confidence intervals for Log2 fold-change estimates alongside p-values. — Confidence intervals convey both the magnitude and precision of an estimated difference; with n = 2 replicates, intervals would be wide and informative about the uncertainty in each estimate, complementing the point estimate and p-value
  • A fixed Log2 fold-change threshold (~3.1) was applied as a co-criterion for significance alongside p < 0.05, but the basis for this threshold was not explained.
    Could also: Combine a statistically derived adjusted p-value threshold (e.g., FDR < 0.05 or 0.10) with a biologically motivated fold-change cutoff stated with explicit justification, or report all genes passing the statistical threshold and use fold-change to rank or annotate them. — Transparently justifying both thresholds—statistical and biological—helps readers understand the sensitivity/specificity trade-off in the gene list and allows comparison across studies using different cutoff conventions
Software: R (TM4 normalization protocol) · MeV (MultiExperiment Viewer) 4.2 · Microsoft Excel

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
18
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.

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.

AAA93296 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
AAC69631 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
AAF72185 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
AAK62673 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
AAM74164 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
AF459093 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
AF515775 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
B90736 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
BAA15794 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
BAB41620 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
BAB83937 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
C89776 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
CAB51807 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
CAD55362 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
G71363 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
G97906 ENA in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
GSE138075 GEO in Data Availability (http://purl.obolibrary.org/obo/IAO_0000611)
no other assessed paper uses this yet
P47768 UniProt in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
P80544 UniProt in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
Q99SD4 UniProt in Results (http://purl.org/orb/Results)
no other assessed paper uses this yet
T44381 ENA in Results (http://purl.org/orb/Results)
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.md — pmid-31681245

Paper: Efthimiou, Tsiamis, Typas, Pappas (2019) Front Microbiol 10:2393. "Transcriptomic Adjustments of Staphylococcus aureus COL (MRSA) Forming Biofilms Under Acidic and Alkaline Conditions." DOI 10.3389/fmicb.2019.02393.

Experiment

Affymetrix GeneChip S. aureus Genome Array (GPL1339), one-colour. S. aureus COL grown at pH 5 / 7 / 9, as biofilm (BF) on nitrocellulose membranes and as planktonic (PL) cells. Biological duplicates (n=2). 12 arrays total (GSE138075). pH 7 is the control in every comparison.

Pipeline as described (Methods + GEO data_processing field)

  1. Raw CEL normalised in R, "TM4 protocol" (http://www.tm4.org/normalizing.html).
  2. Data filtering in MeV (MultiExperiment Viewer Quickstart Guide v4.2), "variance filter value = 50" (under-specified — count? percentile?).
  3. Filtered data exported to Excel; per gene Log2(avg Expr_cond1 / avg Expr_cond2).
  4. Two-tailed paired t-test (Excel) on the per-gene expression values of the two conditions; significant if p < 0.05.
  5. Up/down lists thresholded at log2FC ≈ ±3.1 (KEGG/Aureowiki only for annotation).

The "code" link in the registry is github.com/dfci-cccb/www.tm4.org = the TM4/MeV suite itself, i.e. a third-party tool (P16 case), not author analysis code. There is no author script; the analysis is the generic TM4→MeV→Excel workflow above.

IN SCOPE (pipeline-derived, reproduced here)

The reported result is a set of differentially expressed gene counts and per-gene log2FC + p-values (Tables 1–4) produced by steps 3–4 applied to the normalised data. GEO ships the authors' normalised value matrix (GSE138075_series_matrix.txt.gz, 3887 probesets × 12 arrays) — this is exactly the input to their Excel step. So the faithful 1:1 reproduction is:

  • R1 (core). Recompute, on the authors' deposited normalised values, for each of the 4 comparisons (BF/PL × pH9/pH5 vs pH7): per-probeset log2(mean ratio) + two-tailed paired t-test, then apply the paper's thresholds and the MeV variance filter (tested under several interpretations of "value 50"), and compare the resulting up/down DEG counts to C1–C8.
  • R2. Map the named genes (codY, mecA, sceD, femA, sarA, hfq, …) to array probesets via the GPL1339 annotation and compare per-gene log2FC + p (C9–C14).
  • R3 (independent cross-check, harder). Re-normalise the 12 raw CEL files from scratch (RMA/affy) and repeat R1, to test robustness to the normalisation choice. Blocker: no Bioconductor CDF package exists for this array (saureuscdf absent); needs the Affymetrix S_aureus.CDF (Thermo, registration) + makecdfenv. Attempted as a bonus; documented if blocked.

OUT OF SCOPE (not pipeline / not attempted)

  • Wet-lab: bacterial growth, CFU counts, biofilm crystal-violet assays, RT-PCR validation, microscopy (Figs of biofilm morphology). Manual/experimental.
  • KEGG/Aureowiki functional interpretation narrative (manual annotation).
  • The MeV variance-filter exact semantics are not fully specified by the paper; R1 brackets the plausible interpretations rather than guessing one.

Data / compute

  • Data + all intermediates on «infra»: «path»
  • Compute on «our HPC» via SLURM. Analysis is light (t-tests on 3887×12) but run as a job for provenance. «host» holds only small result tables + this scope.
Figures / tables: TableTables
C16
Reported
all 92 per-gene log2FC values in Tables 1A-4B (as printed)
Reproduced
92/92 match a deposited-data probeset to <=0.05 (most <=0.01); 84/92 also match the printed p to <=0.001
exact
C9
Reported
codY log2FC 3.35, p 0.0043 (Table 1A)
Reproduced
3.3519 / 0.00431 (probe sa_c9313s8155_a_at)
exact
C10
Reported
mecA log2FC 3.32, p 0.0047 (Table 1A)
Reproduced
3.3235 / 0.004743 (probe sa_c9104s7978_a_at)
exact
C12
Reported
femA log2FC 4.12, p 0.0356 (Table 1B)
Reproduced
4.1187 / 0.03555 (probe sa_c9776s8518_a_at)
exact
C13
Reported
sarA log2FC -3.43, p 0.0317 (Table 3B)
Reproduced
-3.4252 / 0.03167 (probe sa_c7899s6890_at; same probe reproduces sarA -3.39 in PL_pH5)
exact
C14
Reported
hfq log2FC -3.53, p 0.0024 (Table 4B)
Reproduced
-3.5346 / 0.002359 (probe sa_c1514s1292_at)
exact
C1
Reported
8 up-regulated genes pH9 biofilm (log2FC>3.16, p<0.05)
Reproduced
177 probesets pass stated thresholds (175 non-control); reported 8 = curated subset
did not match
C5
Reported
4 down-regulated genes pH9 biofilm
Reproduced
167 probesets pass stated thresholds
did not match
C2-C8
Reported
DEG counts 11/16/16 up, 18/11/12 down (other comparisons)
Reproduced
143-196 probesets per comparison pass stated thresholds; reported lists are curated subsets
did not match
C15
Reported
Affy S. aureus Genome Array (GPL1339), 12 arrays, n=2/condition
Reproduced
confirmed from GEO: 12 arrays, 3pH x 2mode x 2rep
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 73/100

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: Q7 · Core claim 🟡
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 +5

Every reported per-gene number reproduces essentially 1:1 from the authors' own GEO-deposited normalised matrix (GSE138075): all 92 log2FC values match a probeset to <=0.05 and the named genes (codY, mecA, ctsR, sceD, femA, agrB, sarA, hfq) match on both log2FC and p — so no fabrication, and the input data is identical. The substantive deviation is in the DEG counts: the stated thresholds (log2FC>~3.1 & p<0.05) pass 143-196 probesets per comparison (177 up / 167 down for pH9 biofilm) versus the reported 8/4, because an under-specified MeV variance filter + manual KEGG/Aureowiki curation shrinks the list. This is authors'-side selective reporting / under-specified method, not a computation error on our side — the listed values are genuine but the lists are a hand-picked subset. Overall yellow: reliable gene-level results with an explainable, authors-side discrepancy in the reported gene-set sizes.

🤝
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.

226.3 k
tokens (I/O) · 18.5 M incl. cache
36 min
runtime · 0.01 CPU-h
1.9 GB
peak RAM
1
HPC jobs
hummel
machine