Datasets
Natural and generated sequence-fragment collections derived from CAMI2 and GTDB references.
Task
Natural vs artificial microbial genome sequence
This GenomeOcean task distinguishes natural microbial sequence fragments from model-generated fragments. It measures discrimination under the study’s dataset construction.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2Overview
Metrics
Table 2 reports classification measures including F1; each preserved result keeps its original metric and unit.
Allowed inputs
DNA sequence representations and the natural/generated class label.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2Evaluation procedure diagram
How it worksConceptual assessment outline
Conceptual overview of the published statistical assessment.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2limited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Results
Each comparison retains its reviewed evaluation scope, dataset and metric. Results are shown without a pooled ranking.
Natural versus GenomeOcean-generated DNA classification · Table 2:
Precision (percent) · Higher values are better.
Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence) · GenomeOcean natural/artificial sequence test
Evidence origin: Independent external evaluation, Author-reported evaluation.
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Table 2:: Precision, Natural versus GenomeOcean-generated DNA classification- Self-generator recognition is not generic biological sequence quality.
- Counts are cohort sizes, not verified per-method successful-prediction coverage.
Comparison details and limitations
DNABERT 2 and NTv 2 standard fine-tuning; GenomeOcean LoRA. Negatives generated by GenomeOcean itself. CAMI 2 train 18,000/validation 2,000; GTDB test 20,000; balanced natural/artificial 2 kb sequences.
- No interval assigned unless printed in source cell.
Automated source review: 2026-09-17. Numerical source review does not establish independent reproduction.
Dots show point estimates. Whiskers show only explicitly defined uncertainty (standard deviation, standard error or a labelled interval); their definitions remain in Table. Unresolved uncertainty is not plotted. Differences do not establish statistical significance.
Showing 3 of 3 matching rows.
Tested configuration
0255075100
Reported scoreMethods and evaluation design
Procedure, tasks and evaluated configurations
How it works
What the evaluation establishes
The study uses CAMI2-derived examples for training and validation and GTDB-derived examples for testing. Natural and generated examples form separate classes. This source separation defines the assessment; a discrimination score is not experimental evidence that a generated sequence has biological function.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2Evaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
These source-backed links do not make different protocols or scores interchangeable.
Run this benchmark
Choose a concrete protocol before running an evaluation. Its inputs, split and scoring rules determine which results can be compared.
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
- Using different reference collections for fitting and testing makes the intended dataset transfer explicit.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2
Limitations and conditions
- A discriminator can learn properties of collection or generation provenance. Its score does not validate a generated genome, pathway or phenotype.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2
Profile review details
Inspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review.
Stable record: reported-task-9f9ab0090f6522Specifications
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 |
|---|---|
| Entity type | Paper-specific evaluation task; this profile is a descriptive evidence summary.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Datasets | Natural and generated sequence-fragment collections derived from CAMI2 and GTDB references.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Organisms | Microbial reference sequence collections; organism membership follows the named source datasets.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Assays | Computational class labels for sequence origin, without a functional measurement in this particular endpoint.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Splits | CAMI2-derived training/validation and GTDB-derived test collections are kept distinct. Exact membership and reference releases remain necessary for reproducibility.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Allowed inputs | DNA sequence representations and the natural/generated class label.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Adaptation | The retained comparison assesses representations through the paper’s classification task; generation and classification are distinct stages.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Metrics | Table 2 reports classification measures including F1; each preserved result keeps its original metric and unit.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
| Baselines | Table 2 includes GenomeOcean and DNABERT-2 representation-based comparators. Their classification results do not by themselves assess sequence function.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods 4.2.5 Generated Sequence Discrimination; Table 2 |
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-source discovery and table/protocol screening; source checked is not independently reproduced. Raw acquisitions not automatically numerical publication approval.
| Paper or primary resource | Version | Reference |
|---|---|---|
| GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies | preprint archived 2025-02-05 | Read source DOI: 10.1101/2025.01.30.635558 |
Historical gaps recorded on 2026-09-17
The catalogue now holds 9 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.
- Independent batch review before import; preserve existing observation identities.
Search and extraction details
complete comparison tables extracted pending publication review
Searches
- GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies primary paper benchmark results
Evidence locations
- Table2; detecting artificial sequences Methods
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.
16 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual overview of the published statistical assessment. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Diagram steps
| GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Conceptual assessment outline Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Entity type Paper-specific evaluation task; this profile is a descriptive evidence summary. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Datasets Natural and generated sequence-fragment collections derived from CAMI2 and GTDB references. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Organisms Microbial reference sequence collections; organism membership follows the named source datasets. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Assays Computational class labels for sequence origin, without a functional measurement in this particular endpoint. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Splits CAMI2-derived training/validation and GTDB-derived test collections are kept distinct. Exact membership and reference releases remain necessary for reproducibility. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Allowed inputs DNA sequence representations and the natural/generated class label. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Adaptation The retained comparison assesses representations through the paper’s classification task; generation and classification are distinct stages. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods 4.2.5 Generated Sequence Discrimination; Table 2 Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsInspected the cited primary abstract and descriptive computational-evaluation sections. Review covers the descriptive claims shown; no executable protocol was reconstructed. Original numerical records retain their prior transcription review. 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
- GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Original source · preprint archived 2025-02-05
- GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Original source · preprint archived 2025-02-05
Technical metadata and extraction receipts
Stable ID: reported-task-9f9ab0090f6522
- areas
- microbes-communities
- tasks
- Natural vs artificial microbial genome sequence
- entity level
- task
- version
- Not reported
- task
- Natural vs artificial microbial genome sequence
- scope note
- Paper-specific evaluation task; protocol completeness requires further extraction.
- comparison panels
- id: part2-genomeocean-2025-T2-d2237c539e; title: Natural versus GenomeOcean-generated DNA classification · Table 2:; protocol id: paper-protocol-a8c67b393443253db5; dataset id: reported-dataset-0c3ac7efe99c37; metric: Precision; unit: percent; direction: higher; result ids: paper-result-da2786542bcec5cf7f; paper-result-fae558ea3909a73b15; paper-result-af3d9eb456d1404344; source ids: part2-genomeocean-2025; source locator: Table 2:: Precision, Natural versus GenomeOcean-generated DNA classification; context: DNABERT 2 and NTv 2 standard fine-tuning; GenomeOcean LoRA. Negatives generated by GenomeOcean itself. CAMI 2 train 18,000/validation 2,000; GTDB test 20,000; balanced natural/artificial 2 kb sequences.; caveats: Self-generator recognition is not generic biological sequence quality.; Counts are cohort sizes, not verified per-method successful-prediction coverage.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17; id: part2-genomeocean-2025-T2-3c62335b1d; title: Natural versus GenomeOcean-generated DNA classification · Table 2:; protocol id: paper-protocol-a8c67b393443253db5; dataset id: reported-dataset-0c3ac7efe99c37; metric: Recall; unit: percent; direction: higher; result ids: paper-result-415548dbb583463a98; paper-result-b3378778c23879d6db; paper-result-f356ae1c7c4122384a; source ids: part2-genomeocean-2025; source locator: Table 2:: Recall, Natural versus GenomeOcean-generated DNA classification; context: DNABERT 2 and NTv 2 standard fine-tuning; GenomeOcean LoRA. Negatives generated by GenomeOcean itself. CAMI 2 train 18,000/validation 2,000; GTDB test 20,000; balanced natural/artificial 2 kb sequences.; caveats: Self-generator recognition is not generic biological sequence quality.; Counts are cohort sizes, not verified per-method successful-prediction coverage.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17; id: part2-genomeocean-2025-T2-92317bff43; title: Natural versus GenomeOcean-generated DNA classification · Table 2:; protocol id: paper-protocol-a8c67b393443253db5; dataset id: reported-dataset-0c3ac7efe99c37; metric: F1; unit: %; direction: higher; result ids: lit-b3-028; paper-result-0209741f5d0e66b495; lit-b3-027; source ids: part2-genomeocean-2025; source locator: Table 2:: F1, Natural versus GenomeOcean-generated DNA classification; context: DNABERT 2 and NTv 2 standard fine-tuning; GenomeOcean LoRA. Negatives generated by GenomeOcean itself. CAMI 2 train 18,000/validation 2,000; GTDB test 20,000; balanced natural/artificial 2 kb sequences.; caveats: Self-generator recognition is not generic biological sequence quality.; Counts are cohort sizes, not verified per-method successful-prediction coverage.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17
- benchmark research
- review date: 2026-09-17; status: complete_comparison_tables_extracted_pending_publication_review; primary sources: part2-genomeocean-2025; inspected locators: Table2; detecting artificial sequences Methods; searched queries: GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies primary paper benchmark results; gaps: Independent batch review before import; preserve existing observation identities.; claim scope: Primary-source discovery and table/protocol screening; source checked is not independently reproduced. Raw acquisitions not automatically numerical publication approval.
- historical missing metadata
- protocol version: not_reported_in_legacy_extract; split: 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: genomeocean-2025; source locator: Methods 4.2.5 Generated Sequence Discrimination; Table 2; 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.