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

Complete Genome Sequencing of Lactobacillus plantarum ZLP001, a Potential Probiotic That Enhances Intestinal Epithelial Barrier Function and Defense Agai

Front Physiol · 2018
L1 68/100 3/4
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: 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: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q4 · Cause of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q2 · Endpoint comparability 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +6
✓ What held up
  • Same input data as the authors
  • Reported values are derivable from the shared data
What did not (or only partly)
  • 🟡Reported values were only indirectly comparable
  • 🟡A deviation arose in the data or preprocessing
  • 🟡A deviation was attributed to the published material
  • 🟡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
68/100
Reproducibility score
0.3 SD below mean
vs. all fields · 1173 studies
🎯 Scores higher than 32% of all assessed papers rank 765 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 core result, partially for the rest. Reproduced 1:1 the genome-level pipeline claims by de-novo assembling the paper's own PacBio reads (SRR8079316) with Flye: chromosome size 3,164,344 bp vs reported 3,164,369 (Δ25 bp), GC 44.66% vs 44.65%, chromosome rRNA 16/16 exact, tRNA 73 vs 69, subread count 73,238 exact. Chromosome size+GC independently CONFIRMED genuine against the deposited assembly GCA_003076435.1/CP021086.1 (3,164,369 bp, 44.66%) -> not fabricated. Two honest discrepancies flagged: (1) paper's 'median subread length 8,661 bp' is actually the MEAN (true median 9,289) - a mislabel, value real; (2) Flye from PacBio-only recovered 5 of 7 plasmids (missed the 67.8 kb plasmid A + one ~15 kb), so total CDS 3,131 vs 3,264 and plasmid GC differ - an assembly-completeness limit of our scoped pipeline, not a paper error. NOT ATTEMPTED (hard 20%): the named code's RAxML phylogeny on 553 single-copy orthologs - the paper specifies neither the comparison genomes nor ortholog-calling parameters, so inputs are unspecified and it is not reproducible from the methods as written; also the 16S MEGA tree and KEGG/COG counts. Did NOT run the Illumina hybrid-polish/closure step.

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 68
    assessed: 2026-06-16 ⛓ ffaebfab3389
✎ 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 2026-06-16
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: opus
Founding hypothesis

What genomic features underlie the probiotic and gut-health-promoting properties of Lactobacillus plantarum ZLP001, a strain isolated from healthy weaned piglet gut, and how does its genome compare with other L. plantarum strains?

Core claims
  • The complete genome of L. plantarum ZLP001 comprises a single 3,164,369 bp circular chromosome (GC 44.65%) plus seven plasmids (A–G), encoding 3,264 protein-coding sequences. resource
  • ZLP001 carries genes (Na+:H+ antiporter, choloylglycine hydrolase, heat shock proteins, chaperones) supporting tolerance to low pH, bile salt, and stress in the gastrointestinal environment. finding
  • ZLP001 harbors an expanded antioxidative gene repertoire (glutathione, thioredoxin, catalase, NADH oxidase/peroxidase systems) but lacks superoxide dismutase, consistent with its high antioxidant ability. mechanism
  • ZLP001 contains 119 CAZyme genes across five families, more than L. plantarum KLDS1.0391, suggesting probiotic potential for pathogen defense and immune stimulation. finding
  • Phylogenomic analysis of 19 L. plantarum strains places ZLP001 close to BDGP2, JDM1, and LZ95 but on a relatively standalone branch, indicating distinct genomic adaptation to the gut. finding
  • The 19 L. plantarum genomes have an open pan-genome of 6,598 orthologous gene families and a core genome of 596 families, with 65 genes unique to ZLP001. finding
  • The genome was sequenced de novo using PacBio SMRT single-molecule real-time sequencing followed by assembly and multi-database functional annotation. method
  • ZLP001 possesses a transport/secretion repertoire of 306 transport genes (PTS, ABC), a Sec-SRP system, two intact prophages, and 22 CRISPR loci. finding
Experimental setups
Assay System Perturbation Readout Platform
Whole-genome sequencing (SMRT/PacBio) Lactobacillus plantarum ZLP001 (from weaned piglet gastrointestinal mucosa) none complete genome sequence (chromosome + plasmids) PacBio RS II, C4 chemistry, P6 polymerase; 8–12 kb library
De novo genome assembly L. plantarum ZLP001 sequence reads none assembled contigs/genome SOAPdenovo v2.04; SMRT Analysis v2.3.0
Gene prediction and functional annotation L. plantarum ZLP001 genome none CDS counts, COG and KEGG functional category assignments Glimmer v3.02; BLAST vs COG/KEGG
Non-coding RNA and repeat prediction L. plantarum ZLP001 genome none rRNA, tRNA, sRNA, tandem/mini/microsatellite counts rRNAmmer v1.2, tRNAscan v1.23, Rfam v10.1, Tandem Repeat Finder v4.04
Mobile genetic element analysis (prophage/CRISPR) L. plantarum ZLP001 genome none prophage regions, integrases, CRISPR/Cas loci PHAST; MinCED 3
CAZyme annotation L. plantarum ZLP001 genome none carbohydrate-active enzyme gene family counts CAZy database
Comparative/phylogenomic and ortholog clustering analysis 19 L. plantarum complete genomes (ZLP001 + 18 from NCBI) none core/pan-genome size, single-copy orthologous families, phylogenetic trees OrthoMCL v2.0, BLASTP, MCL; MAFFT v7, RAxML, MEGA
Genomic DNA extraction L. plantarum ZLP001 culture (MRS broth, 37°C, 18 h, microaerophilic) none purified total genomic DNA Wizard Genomic DNA Purification Kit (Promega)
Key results
  • Circular chromosome of 3,164,369 bp at 44.65% GC, occupying 83.28% of the genome with 3,104 chromosomal genes (avg 886 bp) 3,164,369 bp; 44.65% GC
  • Seven plasmids (A 67,802; B 48,418; C 31,389; D 27,860; E 16,139; F 15,258; G 13,837 bp) with average GC 42.05% 7 plasmids; avg GC 42.05%
  • Pan-genome of 6,598 orthologous families and core genome of 596 families (9.03%); core constitutes 19.59% of each genome pan 6,598; core 596 (9.03%)
  • 65 genes unique to ZLP001 (including 30 hypothetical proteins); strain-specific genes across 19 strains totaled 2,597 (39.36%) 65 unique; 2,597 (39.36%) strain-specific
  • ZLP001 contains 119 CAZyme genes (18 CE, 13 CBM, 32 GT, 50 GH, 6 AA), more than KLDS1.0391 (14 CE, 21 CBM/CPM, 23 GT, 34 GH, 2 AA) 119 vs 94 genes
  • 1,603 CDSs assigned to 39 KEGG categories; 1,783 CDSs assigned to 20 COG categories 1,603 KEGG; 1,783 COG
  • 306 transport-related genes (54 PTS, 252 ABC), 10 complete PTS EII complexes; two intact prophages; 22 CRISPR loci 306 transport; 22 CRISPR loci
  • ZLP001 16S rRNA shows >99% similarity to other L. plantarum strains; clusters near BDGP2, JDM1, LZ95 on a standalone branch; 553 single-copy orthologous families >99% similarity; 553 single-copy families
Key statistics
  • count 3,164,369 bp chromosome (ZLP001 chromosome size)
  • count 3,264 protein-coding sequences (total CDSs in ZLP001 genome)
  • other 44.65% GC (chromosome GC content)
  • count 6,598 orthologous gene families (pan-genome) (pan-genome across 19 L. plantarum strains)
  • count 596 (9.03%) orthologous families (core-genome) (core genome across 19 strains)
  • correlation >99% 16S rRNA gene similarity (ZLP001 vs other L. plantarum strains)
  • count 73,238 subreads, median length 8,661 bp (PacBio sequencing output used for assembly)
  • count 119 CAZyme genes (five CAZyme families in ZLP001 genome)

Statistical methods review

Model: opus

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 is a genome data report describing the complete genome sequencing, assembly, annotation, and comparative/phylogenomic analysis of Lactobacillus plantarum ZLP001 against 18 other publicly available L. plantarum genomes. The work is descriptive and bioinformatic rather than experimental: results are reported as genome counts, gene-family memberships (core/pan/strain-specific), and phylogenetic relationships, with no inferential hypothesis testing of group differences. Tree reliability is the only place a statistical/resampling procedure is reported, via bootstrap support values.

Replicationunclear Sample sizeCounts are stated as numbers of genomes analyzed (1 newly sequenced strain plus 18 from NCBI, 19 total) and sequencing depth (73,238 subreads, median length 8,661 bp); no statistical sample-size or power calculation is described GroupsL. plantarum ZLP001 vs. 18 other L. plantarum genomes (comparative genomics) Pairingna Randomization/blindingna Dispersionnone Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
Neighbor-joining phylogenetic reconstruction with bootstrap resampling (1,000 replicates) 16S rRNA gene tree of L. plantarum and other Lactobacillus (Figure 1C) 1,000 bootstrap replications not stated
Maximum-likelihood phylogenomic reconstruction (RAxML) with bootstrap resampling (1,000 replicates) Concatenated single-copy orthologous gene-family tree of 19 strains (Figure 1D) 1,000 bootstrap repetitions not stated
BLASTP homology search with cutoffs (E-value 1e−5, percent match ≥ 50%) followed by MCL clustering (inflation 1.5) OrthoMCL ortholog clustering; core/pan-genome and strain-specific gene determination (Figure 1E, Tables S1–S2) 19 genomes na
Approaches that could also have been used
  • Genome assembly was performed with SOAPdenovo, a short-read de Bruijn graph assembler, applied to PacBio SMRT long-read subreads.
    Could also: Long-read-oriented assemblers such as HGAP/Canu/Flye could also have been used. — Long-read assemblers are designed around the error profile and length distribution of SMRT data and would also report assembly completeness/contiguity metrics, which can complement a short-read assembler choice.
  • Phylogenetic relationships were summarized using bootstrap support from 1,000 replicates on neighbor-joining (16S) and maximum-likelihood (concatenated orthologs) trees.
    Could also: Bayesian inference (e.g., MrBayes/BEAST) with posterior probabilities, or additional approximate-likelihood-ratio (aLRT/SH-like) support, could also have been reported. — A second support metric or Bayesian posterior would also describe branch confidence under a different statistical framework and is commonly reported alongside bootstrap values.
  • Genome relatedness among strains was assessed via 16S rRNA similarity and concatenated single-copy ortholog phylogeny.
    Could also: Whole-genome distance metrics such as average nucleotide identity (ANI) or digital DNA–DNA hybridization could also have been computed. — ANI/dDDH provide quantitative pairwise genome similarity values that also help resolve closely related strains where 16S has limited discriminatory power, as the paper itself notes for L. plantarum.
  • Core- and pan-genome sizes were reported as single point counts from OrthoMCL clustering of 19 genomes.
    Could also: Pan-genome rarefaction/accumulation curves with fitted models (e.g., Heaps' law) could also have been presented. — Accumulation curves would also describe how core and pan-genome estimates change with the number of sampled genomes and characterize the open/closed nature of the pan-genome quantitatively.
  • Ortholog families were defined using fixed BLASTP cutoffs (E-value 1e−5, ≥50% match) and an MCL inflation of 1.5.
    Could also: A sensitivity analysis varying the inflation value and identity/coverage thresholds could also have been included. — Reporting how core/pan/strain-specific counts respond to parameter choices would also convey the robustness of the clustering results to the chosen thresholds.
Software: SMRT Analysis (PacBio data filtering) v2.3.0 · SOAPdenovo (de novo assembly) v2.04 · Glimmer (gene prediction) v3.02 · Tandem Repeat Finder v4.04 · rRNAmmer v1.2 · tRNAscan v1.23 · Rfam v10.1 · Circos (genome visualization) v0.69-6 · PHAST (prophage prediction) · MinCED (CRISPR prediction) 3 · OrthoMCL (ortholog clustering) v2.0 · MCL (clustering) · MAFFT (alignment) v7 · RAxML (maximum-likelihood trees) · MEGA (neighbor-joining tree)

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
64
Impact: high
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.

AL935263 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP001617 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP002222 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP004082 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP005942 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP006033 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP009236 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP010528 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP012122 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP012650 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP014228 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP015857 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP015966 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP019348 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP021086 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP021997 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP023174 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP023728 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
CP028977 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
PRJNA381357 BioProject in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet
SRP102895 ENA in Methods (http://purl.org/orb/Methods)
no other assessed paper uses this yet

What was reproduced

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

Reproduction scope — pmid-30542296

Paper: Zhang et al. 2018, Complete Genome Sequencing of Lactobacillus plantarum ZLP001, a Potential Probiotic... Front Physiol 9:1689. PMCID PMC6277807. Type: complete-genome announcement.

Data: BioProject PRJNA381357

  • SRR8079316 — PacBio RS II SMRT subreads (73,238 reads, 634 Mb) → assembly input
  • SRR5407012 — Illumina HiSeq 4000 PE (1.52 M reads) → used by authors for polishing
  • Deposited assembly GCA_003076435.1 (ASM307643v1) → independent cross-check

Code link in brief: github.com/stamatak/standard-RAxML (third-party tool the authors used for the phylogeny; not their own pipeline). Per P16 this is valid to reproduce, but see out-of-scope note below.

IN SCOPE — clearly-specified pipeline outputs (80%)

Pipeline: PacBio reads → de-novo assembly (paper: SOAPdenovo v2.04; we use Flye, the standard long-read assembler — SOAPdenovo is a short-read assembler and is not appropriate for PacBio data, so an identical tool match is neither possible nor sensible; genome size/GC are assembler-independent) → gene/RNA annotation (paper: Glimmer v3.02; we use Prodigal/barrnap/tRNAscan-SE).

id result reported robustness
c1 PacBio subread count 73,238 exact, tool-free
c2 PacBio median subread length 8,661 bp exact, tool-free
c3 chromosome size 3,164,369 bp high (assembler-independent)
c4 chromosome GC 44.65% very high (sequence-intrinsic)
c9 plasmid count 7 high
c10-11 plasmid sizes A–G 67,802 … 13,837 bp high
c12 plasmid mean GC 42.05% very high
c5/c6 CDS / chromosome genes 3,264 / 3,104 LOW — gene-caller dependent (Glimmer vs Prodigal differ)
c7/c8 chromosome rRNA / tRNA 16 / 69 medium — tool dependent

The strongest 1:1 evidence is c1–c4, c9–c12 (sequence-intrinsic). CDS/RNA counts are reported with an explicit tool-difference caveat — agreement within a few % is expected, not exact.

OUT OF SCOPE — the hard ~20% (not attempted, with reason)

  • RAxML ML phylogeny on 553 single-copy orthologs (Figure): requires the set of other genomes compared against, which the paper does not enumerate (no species list, no accessions, no ortholog-calling pipeline parameters). Not reproducible from the described methods → skipped per 80/20 rule. The named code link (standard-RAxML) is only the tree-builder; the unspecified inputs are the blocker, not the tool.
  • 16S rRNA neighbor-joining tree (MEGA) — same input-set ambiguity; wet-lab/manual.
  • KEGG/COG category counts (1,603 / 1,783 CDS) — depend on the exact CDS set + database versions (2018); reproducible in principle but low-value and version-sensitive → not attempted.
  • All wet-lab assays (barrier function, defense) — out of computational scope.

Verdict shape expected

A partial reproduction: strong 1:1 on genome size/GC/plasmid structure (sequence-intrinsic, anti-fabrication cross-checked against the deposited assembly), with annotation counts reported as tool-caveated approximations. The phylogeny — the result tied to the named code — is not reproducible from the paper as written.

c1
Reported
73238 PacBio subreads
Reproduced
73238
exact
c2
Reported
subread median length 8661 bp
Reproduced
mean 8661.7 bp (true median 9289)
within tolerance
c3
Reported
chromosome 3164369 bp
Reproduced
3164344 bp (Flye); 3164369 bp deposited
within tolerance
c4
Reported
chromosome GC 44.65%
Reproduced
44.66%
within tolerance
c5
Reported
3264 total CDS
Reproduced
3131 (Prodigal)
partial
c7
Reported
16 chromosome rRNA
Reproduced
16
exact
c8
Reported
69 chromosome tRNA
Reproduced
73
within tolerance
c9
Reported
7 plasmids
Reproduced
5 circular plasmids
partial
c10
Reported
plasmid A 67802 bp
Reproduced
not recovered (PacBio-only)
did not match
c11
Reported
plasmids B-G 48418;31389;27860;16139;15258;13837
Reproduced
45809;32413;30392;27862;13885 (D delta 2bp, G delta 48bp)
partial
c12
Reported
plasmid mean GC 42.05%
Reproduced
39.44% over 5 plasmids
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 68/100

An automated assessment. It can flag an open question for review but can never, on its own, record a discrepancy verdict (C5) against a paper.

🟢1. Data identity
🟡2. Endpoint comparability
🟡3. Location of the main deviation
🟡4. Cause of the deviation
🟢5. Derivability / plausibility
🟡6. Severity of the deviation
🟡7. Core claim
🟡8. Severity of the miss (overall human judgment)
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
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: Q4 · Cause of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q2 · Endpoint comparability 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +6

The central, deposited genome claims reproduce essentially 1:1 — chromosome 3,164,344 bp vs reported 3,164,369, GC 44.66% vs 44.65%, rRNA 16/16, and 73,238 subreads exact — and independently match GenBank CP021086.1, confirming the values are genuine, not fabricated. The remaining deviations (5/7 plasmids, total CDS 3,131 vs 3,264, plasmid mean GC 39.44 vs 42.05) lie on our side: we deliberately skipped the authors' Illumina hybrid closure and used Prodigal instead of Glimmer. The only authors-side issue is a minor mislabel of the mean subread length (8,661 bp) as the median. Overall a solid partial reproduction with fully explainable, non-critical discrepancies.

🤝
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.

93.2 k
tokens (I/O) · 5.2 M incl. cache
18 min
runtime · 0.69 CPU-h
16.1 GB
peak RAM
1
HPC jobs
hummel
machine