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

ddRAD-seq reveals the genetic structure and detects signals of selection in Italian brown trout.

Genet Sel Evol · 2022
L1 87/100 PQI 96
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: 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: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q4 · Cause of the deviation 🟡
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 were directly comparable
  • Any deviation was negligible
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 central claim did not (fully) hold under reproduction
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
87/100
Reproducibility score
0.7 SD above mean
vs. all fields · 1173 studies
🎯 Scores higher than 72% of all assessed papers rank 301 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 -> clean 1:1. The authors deposited the exact STACKS genotype VCF (figshare doi:10.6084/m9.figshare.12999725, the central pipeline input) plus their R/shell repo (github.com/genomeud/GenSal @4f7ea79). Recomputing the clearly-specified downstream numbers directly from that VCF: 90 individuals (exact), 40 chromosomes (exact), 126,134 SNPs vs reported 126,124 (within-tol, diff 10 = 0.008%), and genome-wide FIS 0.3936 vs reported ~0.39 (exact, via the repo's own b11/vcftools --hardy logic). NOT attempted: (a) re-running STACKS from raw reads PRJNA663991 (upstream demux/catalog step is not in the deposited repo) since the deposited VCF makes it unnecessary for these claims; (b) the ZHp selection-region COUNT (17 regions) because a05_ZHp.r leaves the significance threshold commented out and never sets it, and the window->region merging is unspecified -- the Hp/Z values are computable but the count is not cleanly reproducible (the 20%, skipped). Note: VCF header says Stacks v2.4 while Methods say v2.0; BRIEF's data accession PRJEB32115 is the reference genome fSalTru1, not the reads.

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 87
    assessed: 2026-06-15 ⛓ 0362c0cfdef2
✎ 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

What is the genetic structure of Italian brown trout populations, to what extent have stocking practices introduced admixture (particularly with the Atlantic lineage), and can genome-wide ddRAD-seq data detect signatures of selection distinguishing the lineages?

Core claims
  • Italian brown trout populations are genetically differentiated but show strong admixture introduced by stocking, especially with the Atlantic lineage. finding
  • Most analysed populations show high levels of kinship and inbreeding, indicating eroded genetic diversity. finding
  • Regions putatively under selection with reduced heterozygosity across populations are enriched for genes involved in the response to viral infections. finding
  • ddRAD-seq generated more than 100,000 SNPs in 90 brown trout samples to characterize population structure and selection signatures. resource
  • Putative signatures of selection were detected using complementary approaches (ZHp pooled heterozygosity, HapFLK haplotype-based, IBD sliding-window kinship). method
  • Candidate selected regions associated with resistance to infectious diseases constitute candidates for studying infection resistance in wild and farmed trout. finding
  • The Mediterranea lineage was subdivided into Mediterranea Island and Mediterranea Mainland, yielding five lineages. finding
Experimental setups
Assay System Perturbation Readout Platform
ddRAD-seq (double-digest restriction-site associated DNA sequencing) genotyping Salmo trutta (brown trout); fish farms and rivers in Italy, Corsica, Austria none >100,000 SNP genotypes Illumina HiSeq2500, paired-end 2×125 bp, V4 chemistry; SphI/BstYI digestion; BluePippin size selection; MagAttract HMW DNA kit
Read alignment and variant calling Salmo trutta v1.1 reference genome (NCBI PRJEB32115) none loci detection and genotyping Stacks v2.0 (process_radtags, gstacks, populations); BWA-MEM
Population structure / admixture analysis 90 brown trout samples (5 lineages) none ancestry probabilities Q, K populations Admixture software
Hybrid characterization subset of SNPs (high FST >0.95, low LD <0.2) none hybrid class assignment NewHybrids
PCA and phylogenetic tree analysis brown trout SNP dataset none principal components, ML trees with migration events R SNPRelate; SNPhylo; iTOL; Treemix
Relatedness/IBD and inbreeding estimation brown trout individuals within and between lineages none IBD, inbreeding coefficient FIS = 1-Ho/He, heterozygosity, nucleotide diversity π R SNPRelate
Signatures of selection — pooled heterozygosity five lineages (whole sample and per population; farmed vs wild) none ZHp Z-score in 1 Mb sliding windows (step 200 kb, ~67 SNPs)
Haplotype-based selection scan studied brown trout populations none haplotype divergence p-values (windows of ≥2 consecutive SNPs) HapFLK (K=20, 20 iterations)
Key results
  • Stocking practices introduced strong admixture in endemic Italian trout, especially with the Atlantic lineage.
  • Most analysed populations showed high levels of kinship and inbreeding.
  • Regions under selection with reduced heterozygosity across all populations are enriched for genes involved in response to viral infections.
  • More than 100,000 SNPs were obtained from 90 brown trout samples via ddRAD-seq. >100,000 SNPs
  • Admixture analysis identified five ancestral populations/lineages (K=5): AT, CA, MA, MI, MM. K=5
Key statistics
  • count >100,000 SNPs (SNPs genotyped across brown trout samples)
  • count 96 fishes sampled (individuals sampled from Italy, Corsica, Austria)
  • count 6 samples excluded (avg coverage <5×) (samples removed from analysis for low coverage)
  • other ZHp threshold < −2.81 or > 2.81 (two-tail p ≤ 0.005) (significance threshold for pooled heterozygosity selection signal)
  • pvalue nominal p ≤ 0.01 (HapFLK putative selected regions (≥2 consecutive SNPs))
  • other maxQ > 0.95 (threshold defining admixed individuals)
  • other kinship coefficient ≥ 0.05 (IBD sliding-window candidate regions (5 Mb window, 2.5 Mb step))
  • other FST > 0.95, LD < 0.2 (SNP subset selection for NewHybrids analysis)

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 ddRAD-seq to genotype 90 brown trout individuals across five Italian lineages, employing ADMIXTURE, PCA, and maximum-likelihood phylogenetic methods (SNPhylo, TreeMix) to characterize population structure and gene flow. Relatedness and inbreeding were estimated with ML-IBD and F_IS via SNPRelate. Putative signatures of selection were detected with three complementary genome-scan approaches—ZHp sliding-window heterozygosity, HapFLK haplotype-based analysis, and sliding-window IBD kinship—each with permutation-based false-positive control (1,000 reshuffles). Candidate selected regions were functionally annotated using KEGG pathway enrichment.

Replicationbiological Sample size96 fish sampled; 6 excluded for mean coverage < 5×, yielding n = 90; no a priori power calculation described GroupsFive lineages: Atlantic (AT), Carpione (CA), Marmoratus (MA), Mediterranea Island (MI), Mediterranea Mainland (MM); secondary stratification by farmed vs. wild origin Pairingunpaired Randomization/blindingnot stated Dispersionnone Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionnone stated; within-scan empirical false-positive control via permutation resampling (1,000 reshuffles, region-level p < 0.05) applied independently to each of the three selection scans; KEGG enrichment uses nominal p < 0.05 with minimum term-count filter only
Statistical tests used
Test Applied to n Assumptions
ADMIXTURE (ML-based ancestry proportion estimation, K determined by cross-validation) Population structure across all five lineages / Fig. 1b 90 not stated
NewHybrids (Bayesian hybrid class assignment) Hybrid structure characterization using SNPs with FST > 0.95 and LD < 0.2 90 not stated
PCA (principal component analysis via SNPRelate) Population structure visualization 90 na
Maximum-likelihood phylogenetic tree with bootstrap (SNPhylo) Phylogenetic relationships among lineages 90 not stated
TreeMix ML phylogenetic network with migration edges (1–6 migration events evaluated) Gene flow directions between lineages 90 not stated
ML estimator of pairwise IBD (SNPRelate); F_IS = 1 − Ho/He Within- and between-lineage relatedness and inbreeding characterization 90 not stated
ZHp statistic (Z-transformed pooled heterozygosity), sliding window (1 Mb window, 200 kb step, ~67 SNPs/window), |ZHp| > 2.81 (two-tail p ≤ 0.005), ≥ 2 overlapping significant windows required; permutation resampling (1,000 reshuffles, region-level p < 0.05) Genome-wide scan for enrichment/depletion of homozygosity as signatures of selection 90 not stated
HapFLK haplotype-based selection test (K = 20 local haplotype clusters, 20 iterations), nominal p ≤ 0.01 for ≥ 2 consecutive SNPs; permutation resampling (1,000 reshuffles, region-level p < 0.05) Genome-wide scan for recent selective sweeps differentiating studied populations 90 not stated
Sliding-window IBD kinship estimation (SNPRelate; 5 Mb window, 2.5 Mb step), average pairwise kinship ≥ 0.05 across all population pairs in ≥ 2 overlapping windows; permutation resampling (1,000 reshuffles, region-level p < 0.05) Identification of IBD-elevated regions shared between populations as a proxy for selection 90 not stated
KEGG pathway enrichment analysis (specific test not stated; p < 0.05, minimum 3 pathway occurrences required to test) Functional annotation of putative selected regions not stated
Approaches that could also have been used
  • Population structure was inferred using ADMIXTURE with cross-validation to choose K; no second structural method was applied as a primary comparison
    Could also: STRUCTURE (MCMC-based) or sNMF could also estimate ancestry proportions; DAPC (discriminant analysis of principal components) provides a model-free complement — Comparing results across two algorithms with different assumptions (EM vs. MCMC; parametric vs. model-free) can confirm that inferred K and Q-values are robust to methodological choices, which is informative given the complex admixture history of these lineages
  • Signatures of selection were scanned with ZHp, HapFLK, and IBD kinship, each controlled by its own within-method permutation at region-level p < 0.05, with no joint FDR across the three methods
    Could also: A single genome-wide Benjamini-Hochberg FDR correction applied jointly across all tested windows (or all SNPs) within each method would also be a standard approach; outlier methods such as PCAdapt or BayPass that model population structure as a nuisance covariate could additionally serve as comparators — A pooled FDR across all windows provides a directly interpretable false-discovery rate; methods that co-model structure reduce confounding between differentiation driven by demography and that driven by selection
  • KEGG pathway enrichment was assessed at nominal p < 0.05 with a minimum term-count filter of 3 and no correction across pathways
    Could also: Benjamini-Hochberg FDR correction across all tested pathways could also be applied; a hypergeometric or Fisher's exact test framework (e.g., clusterProfiler) with FDR adjustment is a widely used alternative — When many pathways are tested simultaneously, applying an FDR correction reduces the expected number of false-positive enriched categories and is commonly reported to aid interpretation
  • Inbreeding was quantified as F_IS = 1 − Ho/He at the SNP level and via the ML-IBD estimator in SNPRelate
    Could also: Runs-of-homozygosity (ROH)-based inbreeding coefficients (F_ROH) could also be estimated from the dense SNP dataset; KING-robust or GCTA GREML could provide additional genomic relatedness matrices — F_ROH captures autozygosity from extended IBD segments and can distinguish recent (long ROH) from ancestral (short ROH) inbreeding, providing a complementary perspective to the single-locus F_IS approach, especially relevant for the small population sizes noted in this study
  • Recombination rates near putative selected regions were approximated by anchoring a Salmo salar linkage map to the S. trutta genome via BLAST, treating S. salar rates as proxies
    Could also: Crossover rates could also be estimated directly from the ddRAD genotype data using LD-decay-based methods such as LDhat or LDhelmet within the S. trutta dataset itself — Direct within-species estimation avoids the assumption that recombination landscapes are conserved between S. salar and S. trutta, which may differ in locally rearranged regions, and would yield recombination estimates for all candidate loci rather than only those with anchored linkage-map markers
  • Farmed and wild individuals were combined in primary analyses and stratified post hoc by repeating key analyses separately in each group
    Could also: A mixed-model framework (e.g., GEMMA or SAIGE) that explicitly includes farm/wild origin as a fixed or random effect in the selection scan could also be applied — Explicit statistical modeling of farm/wild status as a covariate in the genome scan can formally partition domestication-associated signals from signals of natural local adaptation, complementing the stratified replication approach used here
Software: Stacks (process_radtags, gstacks, populations) 2.0 · BWA-MEM · ADMIXTURE · NewHybrids · R / SNPRelate · SNPhylo · iTOL (web tool) · TreeMix · HapFLK · BLAST / blastx · R / KEGGREST (Bioconductor)

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
20
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.

PRJEB32115 BioProject 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.

n_individuals
Reported
90
Reproduced
90
exact
n_chrom
Reported
40
Reproduced
40
exact
snps_total
Reported
126124
Reproduced
126134
within tolerance
fis_genomewide
Reported
~0.39
Reproduced
0.3936
exact
zhp_regions
Reported
17 (13 neg, 4 pos)
Reproduced
not attempted
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 87/100

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: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q4 · Cause of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +6

Strong, near-1:1 reproduction: the authors deposited the exact STACKS genotype VCF and their own pipeline repo, and recomputing directly from that file matches the paper on individuals (90), chromosomes (40), SNP total (126,134 vs 126,124, Δ0.008%) and genome-wide FIS (0.3936 vs ~0.39). The one gap is the ZHp selection-region count (17), which is not derivable from the shared code because the significance threshold is commented out and the window→region merging is unspecified — an authors-side code incompleteness, not a fabrication signal. A minor provenance inconsistency (Stacks v2.4 in the VCF header vs v2.0 in Methods) is noted but immaterial. Overall yellow: solid descriptive reproduction with the selection-scan half of the central claim only partially confirmable.

🤝
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.

90.8 k
tokens (I/O) · 4.3 M incl. cache
9 min
runtime · 0.01 CPU-h
1.3 GB
peak RAM
1
HPC jobs
hummel
machine