Datasets
The curated FusionAI benchmark.
Task
gene fusion breakpoint classification
Fusion-breakpoint classification evaluates curated sequence labels rather than locating breakpoints in raw sequencing data.
Sources
Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90Overview
Metrics
Accuracy, class-weighted precision/recall/F1 and ROC-AUC.
Allowed inputs
Genomic breakpoint sequence represented by foundation-model embeddings.
Sources
Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90Evaluation procedure diagram
How it worksComputational evaluation flow
Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol.
Sources
Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90limited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Results
Results are available, but no reviewed comparison panel is linked in this release.
All evaluations
1 evaluation · 1 result. Different protocols are not a single leaderboard.
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Pipeline: Nucleotide Transformer + NN (middle) | Task: gene fusion breakpoint classification Dataset: gene fusion breakpoint DNA sequences | 0.994 ROC AUC fraction · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourceNucleotide Transformer + NN (middle): gene fusion breakpoint classification middle embedding with neural-network classifier Aggregation: Not reported Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Table 2, NT / NN (middle) row, ROC AUC column |
Source checking is not independent reproduction. Release 2026-09-29-06401fd5b220.
Methods and evaluation design
Procedure, tasks and evaluated configurations
How it works
Evaluation methodology
The curated FusionAI benchmark. Accuracy, class-weighted precision/recall/F1 and ROC-AUC. Foundation-model embeddings with supervised heads are compared with FusionAI.
Sources
Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90; Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90; Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90Recorded evaluations
Each evaluation records what was tested and under which conditions.
Run instructions
No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.
A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.
Strengths, limitations and unresolved questions
Strengths and limitations
Strengths and considerations
No source-reviewed explanatory claims are recorded here yet.
Limitations and conditions
- Breakpoint localization is described as future work. This classification task should not be represented as validated coordinate prediction.
Sources
Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90
Profile review details
Relevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged.
Stable record: reported-task-ee34721cf55590Specifications
Inputs, training, access and other details
Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.
| Property | Description and evidence |
|---|---|
| Datasets | The curated FusionAI benchmark.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Splits | Sampling and partitioning use a fixed random seed so models share the same examples. The reported split is not described as holding out entire genes or breakpoint families.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: implementation/reproducibility, random seed 42; Dataset and Classification |
| Metrics | Accuracy, class-weighted precision/recall/F1 and ROC-AUC.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Baselines | Foundation-model embeddings with supervised heads are compared with FusionAI.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Leakage controls | All model comparisons use the same fixed-seed partitions. This supports matched comparisons but does not by itself prevent related genomic loci crossing the split.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Uncertainty | The paper reports final-epoch neural results and single-run SVM results; repeated-run uncertainty is not established in the reviewed passage.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Entity type | Paper-specific computational evaluation protocol.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Organisms | Human gene fusions from the reused FusionAI classification dataset; the foundation models’ multispecies pretraining coverage is a different property.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Dataset/methods description; Discussion of human gene-fusion classification; cached paragraphs 77,84 |
| Assays | Fusion breakpoint labels.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Allowed inputs | Genomic breakpoint sequence represented by foundation-model embeddings.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
| Adaptation | Supervised prediction heads are compared with FusionAI.SourcesBenchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 |
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Papers and result coverage
Last literature check: 2026-09-17. Primary-paper discovery and source inspection. Source-checked results are not independently reproduced experiments.
| Paper or primary resource | Version | Reference |
|---|---|---|
| Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences | journal full text in PMC | Read source DOI: 10.1186/s13040-026-00553-1 |
Historical gaps recorded on 2026-09-17
The catalogue now holds 1 result rows for this benchmark. A note below about pending extraction describes the state on 2026-09-17 and may since have been answered by a later batch. The result rows and their sources are the current record.
- Counts are approximate (~36k train, ~8k validation, ~8k test) and must remain approximate.
- NN values are final-epoch performance; SVM is a single run. No replicate confidence bounds.
- Do not assign embedding-plus-classifier scores to bare foundation-model checkpoints.
Search and extraction details
primary comparison table screened
Searches
- "PMC13182013"
Evidence locations
- Table 2
- Methods; Evaluation metrics; Implementation details
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
17 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Diagram steps
| Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Computational evaluation flow Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Datasets The curated FusionAI benchmark. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Splits Sampling and partitioning use a fixed random seed so models share the same examples. The reported split is not described as holding out entire genes or breakpoint families. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: implementation/reproducibility, random seed 42; Dataset and Classification Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Adaptation Supervised prediction heads are compared with FusionAI. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Metrics Accuracy, class-weighted precision/recall/F1 and ROC-AUC. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Baselines Foundation-model embeddings with supervised heads are compared with FusionAI. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Leakage controls All model comparisons use the same fixed-seed partitions. This supports matched comparisons but does not by itself prevent related genomic loci crossing the split. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Uncertainty The paper reports final-epoch neural results and single-run SVM results; repeated-run uncertainty is not established in the reviewed passage. Individual claims | Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90 Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
View linked audit checks and correction history
Release 2026-09-29-06401fd5b220 · Record review: needs review
2 source records and release history
- Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Original source · journal full text in PMC
- Benchmarking genomic foundation models for binary classification of gene fusion breakpoints from DNA sequences · Original source · journal full text in PMC
Technical metadata and extraction receipts
Stable ID: reported-task-ee34721cf55590
- areas
- dna-genomes
- tasks
- gene fusion breakpoint classification
- entity level
- task
- version
- Not reported
- task
- gene fusion breakpoint classification
- scope note
- Paper-specific evaluation task; protocol completeness requires further extraction.
- benchmark research
- review date: 2026-09-17; status: primary_comparison_table_screened; primary sources: expansion-p3-fusion-breakpoint-foundation-models-2026; inspected locators: Table 2; Methods; Evaluation metrics; Implementation details; searched queries: "PMC13182013"; gaps: Counts are approximate (~36k train, ~8k validation, ~8k test) and must remain approximate.; NN values are final-epoch performance; SVM is a single run. No replicate confidence bounds.; Do not assign embedding-plus-classifier scores to bare foundation-model checkpoints.; claim scope: Primary-paper discovery and source inspection. Source-checked results are not independently reproduced experiments.
- historical missing metadata
- protocol version: not_reported_in_legacy_extract
- metadata review scope
- historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
- legacy kinds
- benchmark
- entity classification
- review date: 2026-09-17; rationale: This source-scoped record identifies the biological prediction task and holds its paper context. Preserve the existing task identity; exact split, model adaptation and scoring remain in linked evaluations or separate protocol records.; source ids: fusion-breakpoint-foundation-models-2026; source locator: Methods: Evaluation metrics; implementation; Discussion; cached text lines 34–35, 45, 88–90; ambiguities: A paper- or suite-specific task may constrain some inputs or metrics; that alone does not make it interchangeable with a complete versioned protocol. No protocol equivalence is inferred.; Some legacy profile Entity type facts use the generic phrase computational evaluation protocol. That boilerplate is not sufficient to establish a single fixed protocol identity or to merge this task with another protocol record.