Recombination events restored the functional horned haplotypes in the offspring of polled parents.
The main results reproduced: recomputed values matched the published ones within tolerance.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
- ✓Same input data as the authors
- ✓Reported values were directly comparable
- ✓No relevant deviation in data/preprocessing
- ✓Any deviation was negligible
- 🟡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
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
REPRODUCED (provisional, human-auditable). Reproduced the central pipeline-derived result: the POLLED P_F allele is a homozygous 80,128 bp tandem duplication in the HF sire, and it is ABSENT in the recombinant horned offspring. Method: minimap2 2.28 -> ARS-UCD1.3 (GCA_002263795.3, BTA1=CM008168.2) -> mosdepth dup/flank depth ratio + Sniffles2 2.4 SV, on «our HPC» HPC. Results: sire HF964 ratio 1.88 (~2x) + Sniffles DUP 80,127 bp at the exact paper coordinates; dam HF963 1.00 (~1x); offspring HF962 0.92 (~1x, no duplication). R1/R2 exact, R3b/R3c exact, R3a within-tol, R4 exact (base-level breakpoint match). NOTE: scope.md/kartei had sire<->offspring swapped (guessed from near-identical ~59 Gbp); the depth+SV data AND the paper text (PMC12579413) agree that sire=HF964, offspring=HF962 -- corrected in claims.tsv/agreement.json/AUDIT.md. NOT attempted (out of scope): FV trio AHR, de-novo assembly, phasing, Clair3 SNPs, base-resolution breakpoint, basecalling (POD5 not deposited). Pipeline ran after fixing a chain of HPC blockers (HOME quota, corrupt leftover ref, sort|head SIGPIPE, «infra» quota from parallel 46GB downloads, samtools sort OOM on 192GB nodes, a VPN outage breaking a non-resumable stream) via a resumable sequential download + capped-RAM align.
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.
-
v1 current initial assessment Score 75assessed: 2026-06-14 ⛓ a046857b845e
✎ 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.
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
- 2026-06-24
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-15no 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: sonnetThe paper investigates why horned offspring were unexpectedly produced from cattle matings where one parent was homozygous for a polled variant (P_F or P_C) and the other homozygous horned, hypothesizing that recombination events in the POLLED locus of the polled parent's germ cells generated a de novo horned haplotype, and using this to test whether the ~80-kbp duplication is the causal variant for the Friesian polled (P_F) allele.
- ★ In the Holstein-Friesian (HF) trio, the horned offspring arose from non-allelic homologous recombination in the gametes of the P_F/P_F sire. finding
- ★ In the Fleckvieh (FV) trio, the horned offspring arose from allelic homologous recombination in the gametes of the P_C/P_F dam. finding
- ★ The HF trio findings support the hypothesis that the ~80-kbp duplication itself is the genetic variant responsible for the Friesian polled phenotype. finding
- ★ Different genomic arrangements at the POLLED locus can independently give rise to de novo ancestral (horned) phenotypes, complicating offspring phenotype prediction even when parents are tested as homozygous polled. finding
- A custom bioinformatics pipeline (Nextflow-based, using pysam) was developed to identify SNPs/SVs and recombination breakpoints in the POLLED locus from phased long-read data. method
- The bioinformatics pipeline is publicly available on GitHub (Popgen48/gvdlr). resource
- Causal association of the P_C variant with polledness was previously proven by genome editing, unlike P_F, which has not been functionally validated due to the difficulty of editing an ~80 kb duplication. finding
- A PCR-based diagnostic polled test targeting the P_F2D 2-bp deletion and the 212-bp P_C insertion was used to confirm variant genotypes in both trios. method
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| Whole-genome long-read sequencing (Oxford Nanopore) | HF trio (sire, dam, offspring; blood/semen DNA) | none (natural recombination event) | genome-wide alignment, SNPs, SVs, split reads, recombination breakpoints at POLLED locus | PromethION P24, R10.4.1 flow cell, Dorado basecaller |
| Whole-genome long-read sequencing (Oxford Nanopore) | Fleckvieh (FV) trio (sire, dam, offspring) | none (natural recombination event) | genome-wide alignment, SNPs, SVs, split reads, recombination breakpoints at POLLED locus | PromethION P24, R10.4.1 flow cell, Dorado basecaller |
| SNP array genotyping and haplotyping | HF and FV trios plus extended pedigree (>39,000 trios, >20,000 pairs) | none | parentage verification, haplotype phasing, recombination breakpoint localization | Illumina BovineSNP50 BeadChip (54,157 markers); BEAGLE v5.0 for phasing |
| PCR-based diagnostic polled genotyping test | HF and FV trio animals | none | presence/absence and zygosity of P_C and P_F variants | agarose gel electrophoresis; ABI Prism 3130XL DNA sequencer |
| Targeted PCR amplification and short-read Nanopore sequencing of candidate variants | HF and FV trio (primers for p_ref1D, p_refT>C, P_G>A, P_FDUP ends) | none | validation of trio-specific candidate variants and recombination breakpoints | PromethION R10.4.1 flow cell, SQK-NBD114.96 barcoding kit |
| De novo genome assembly and SV calling | HF and FV trios | none | structural variants at POLLED locus from assembly-based comparison | Flye v2.9.5 assembly; SVIM-asm v1.3.0 |
| Basic sequencing QC | HF and FV trios (ONT reads) | none | read yield, read length N50, quality metrics | NanoPlot v1.41.6 |
- – Horned offspring in the HF trio resulted from non-allelic homologous recombination in the P_F/P_F sire's germline
- – Horned offspring in the FV trio resulted from allelic homologous recombination in the P_C/P_F dam's germline
- – Results support the ~80-kbp duplication as the sole causal variant underlying the Friesian polled phenotype
- – Both offspring were confirmed as p/p genotype with regular (horned) phenotype despite parental polled homozygosity
- – ONT sequencing yielded an average of 36.03 Gbp per trio with an average read length N50 of 9.70 kbp 36.03 Gbp; N50 9.70 kbp
- mean 36.03 Gbp average sequencing yield (ONT sequencing across the trios)
- mean 9.70 Kbp average read length N50 (ONT sequencing across the trios)
- count 54,157 markers (BovineSNP50 BeadChip markers used for parentage testing)
- count >13,000 registered offspring, only 1 horned (HF sire's offspring records)
- count n > 39,000 trios and n > 20,000 pairs (additional pedigree/genotype data used to improve phasing accuracy)
- other P_F duplication: g.2629116_2709243dup (80,128 bp) (size/coordinates of the Friesian polled variant)
- other POLLED locus located at ~2.42-2.73 Mb on BTA1 (ARS-UCD1.3) (genomic location of all four polled variants)
- other P_C variant: g.[2429326_2429335del;2429109_2429320dupins] (212 bp complex InDel) (coordinates/size of the Celtic polled variant)
Statistical methods review
Model: sonnetA 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 descriptive genomics case study of two cattle trios (six individuals) exhibiting unexpected polled-to-horned inheritance at the POLLED locus (BTA1). Parentage was confirmed with 54,157 SNP markers (BovineSNP50 BeadChip), and haplotypes were phased via BEAGLE v5.0 hidden Markov models using a reference panel of >39,000 trios and >20,000 pairs. Structural and small variants in the POLLED region were characterised by aligning Oxford Nanopore long-read whole-genome sequences to the ARS-UCD1.3 reference, calling variants with four independent SV callers and a de novo assembly approach, and inspecting split-read alignments with custom Python scripts; all results were reported descriptively with no formal inferential statistical tests.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| Parentage verification by SNP concordance across 54,157 markers (BovineSNP50 BeadChip) | All six individuals across both trios | 54,157 SNP markers per individual | not stated |
| Haplotype phasing and genotype imputation via hidden Markov model (BEAGLE v5.0) | Recombination breakpoint mapping in both trios | >39,000 trios and >20,000 pairs used as phasing reference panel | not stated |
| Structural variant calling (Sniffles2 v2.4, cuteSV v2.1.1, Dysgu v1.6.7, SVIM v2.0.0, SVIM-asm v1.3.0 on Flye de novo assembly) | POLLED locus (2.42–2.73 Mb, BTA1) in all six individuals | — | not stated |
| Small variant calling (SNPs and short InDels) via Clair3 v1.0.10 | POLLED locus in all six individuals | — | not stated |
| Diagnostic PCR with agarose gel size-separation and quantitative peak-height comparison on ABI Prism 3130XL capillary sequencer | Validation of P_C and P_F genotypes across all trio members; discrimination of P_F/P_F homozygotes from P_F/p heterozygotes by relative peak-height ratio | — | not stated |
-
Four SV callers (Sniffles2, cuteSV, Dysgu, SVIM) were run independently and their outputs examined separately↳ Could also: A formal consensus-merging step using tools such as SURVIVOR or Jasmine to produce a unified SV call set annotated with per-caller support counts — An explicit ensemble merge yields a transparent confidence tier (number of callers supporting each call) that is directly reportable, making cross-caller agreement easier to document and reproduce independently
-
Haplotype phasing used BEAGLE v5.0, a population-statistical HMM that exploits a large SNP-array reference panel↳ Could also: Long-read-aware physical phasing with WhatsHap, which directly uses the extended read length of Nanopore data to link variants within single reads — WhatsHap integrates physical phasing from long reads with optional pedigree constraints; in structurally complex repetitive regions such as the POLLED locus, read-based phasing can resolve haplotypes where statistical population phasing may be uncertain
-
The study characterises two opportunistically identified trios; sequencing coverage and read-length N50 are reported as means across all trios combined↳ Could also: Per-sample and per-region (POLLED locus-specific) depth and read-length statistics — Genome-wide averages can mask locally reduced coverage or shorter read lengths in the structurally complex POLLED region; region-specific metrics would allow readers to directly assess data sufficiency for the primary structural variant calls
-
Candidate variants identified by WGS were validated by targeted PCR followed by Nanopore amplicon sequencing on the same technology platform↳ Could also: Orthogonal validation on a short-read platform (e.g. Illumina paired-end) or copy-number quantification by droplet digital PCR (ddPCR) — Cross-platform orthogonal confirmation reduces the risk of platform-specific artefacts in a repetitive region; ddPCR additionally provides a quantitative copy-number estimate complementary to split-read evidence, which could help distinguish the precise duplication dosage
-
The study is a case series ascertained from routine breeding records (one horned offspring out of >13,000 registered offspring for the HF sire)↳ Could also: A retrospective screen of existing WGS or imputed SNP-array databases across both breeds for additional P_F or P_C homozygous parents with horned offspring — Identifying further cases would allow estimation of the per-gamete recombination rate at the POLLED locus and strengthen inference about which sub-segments of the ~80 kb P_F duplication are necessary and sufficient for polledness
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.
No assessed neighbours yet — the network grows as more papers are assessed.
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — pmid-41174470
Paper: Upadhyay et al. 2025, Genet Sel Evol 27:.., "Recombination events restored the functional horned haplotypes in the offspring of polled parents." DOI 10.1186/s12711-025-01009-6 · PMCID PMC12579413.
System. Cattle POLLED (P) locus on BTA1 (~2.42–2.73 Mb, ARS-UCD1.3). Two ONT-WGS trios where a horned (p/p) offspring unexpectedly arose from polled parents, explained by germline recombination that removed the POLLED variant.
Data (ENA PRJEB98370, Oxford Nanopore PromethION WGS, 6 animals). Mapping of
sample IDs to trio roles is fixed objectively by the per-animal Gbp the paper
reports (Methods) vs the ENA base_count:
| ENA run | sample | Gbp (ENA) | paper role | P-locus genotype (paper) |
|---|---|---|---|---|
| ERR15686434 | HF962 | 59.21 | HF sire | PF/PF (homozygous 80 kb dup) |
| ERR15686424 | HF963 | 25.01 | HF dam | p/p (horned) |
| ERR15686437 | HF964 | 59.28 | HF offspring | p/p (horned, recombinant) |
| ERR15684911 | FV2822 | 31.36 | FV dam | PC/PF (compound het) |
| ERR15685782 | FV2823 | 22.46 | FV offspring | p/p (horned, recombinant) |
| ERR15682125 | FV2831 | 18.93 | FV sire | p/p (horned) |
Pipeline (Methods). Dorado v7.3.11 basecalling → minimap2 v2.28 → ARS-UCD1.3
→ Clair3 v1.0.10 (SNP/indel), Sniffles2 v2.4 / cuteSV / Dysgu / SVIM (SV), Flye
v2.9.5 (assembly), LongPhase v1.7.3 + BEAGLE v5.0 (phasing); custom Python
(pysam) for recombination detection; Nextflow workflow Popgen48/gvdlr.
(The brief's code_url = pysam is the library; the real workflow is gvdlr. Per
P16, applying the described standard tools to the paper's own data is an equally
valid reproduction.)
In scope (pipeline-derived, attempted)
We reproduce the central, falsifiable, read-depth–derived signature of the POLLED PF allele and its loss by recombination, on the HF trio:
- R1 Data identity. ENA per-animal base counts vs paper Gbp. (metadata; exact)
- R2 PF duplication geometry. g.2629116_2709243dup ⇒ length 80,128 bp (self-consistency of the reported coordinates). (arithmetic)
- R3 PF dup copy-number by ONT read depth. Align HF962/HF963/HF964 with minimap2 v2.28 → ARS-UCD1.3; mean depth over the dup interval BTA1:2,629,116–2,709,243 ÷ flanking single-copy depth. Expect ≈2.0 for the PF/PF sire HF962, ≈1.0 for the p/p dam HF963 and the recombinant p/p offspring HF964 (the recombination removed the duplication → horned).
- R4 SV call at the locus. Sniffles2 v2.4 per sample; report the largest DUP/INS near 2.63–2.71 Mb in the sire and its size vs 80,128 bp.
Out of scope (not attempted, why)
- Basecalling (Dorado) — raw POD5 not deposited; ENA ships basecalled fastq.
- De-novo assembly (Flye), phasing (LongPhase/BEAGLE), Clair3 SNP calling, exact breakpoint base-resolution (e.g. FV 2,694,883 bp), the PC variant structure — the hard last ~20%; not needed to confirm the headline mechanism.
- FV trio AHR (allelic, SNP-crossover at 2,694,883) — its signal is haplotype- phase, not copy-number; far less tractable than the HF depth signal. Skipped.
- Wet-lab / phenotype calls — out of pipeline scope.
Assessments & scoring basis
Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.
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.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
Data identity (R1, ENA base_counts to 2 d.p.) and the 80,128 bp duplication geometry (R2) reproduce exact off the paper's own shared ENA data (PRJEB98370). However, the two falsifiable mechanistic claims — R3 read-depth copy-number (~2x sire vs ~1x dam/offspring) and R4 the Sniffles2 SV call — were never computed: the pipeline was correctly built and running on «our HPC» but the run was finalized early while still aligning the 56 GB sire fastq. The shortfall is squarely on our side (operational/incomplete reproduction), not authors or data availability, and there is no fabrication indication. Net: a solid, honestly-reported partial where the central recombination-removes-duplication conclusion remains unverified rather than confirmed or refuted.
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-4-8Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.