Model type
Study-specific predictive method; this record is the paper-specific evaluated configuration.
Configuration
GenomeOcean
GenomeOcean is a generative genome model trained on metagenomic contigs spanning diverse environments.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Introduction (paragraph 5); Abstract (paragraph 1)How it worksEvaluated procedure (conceptual)
Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1)At a glance
Biological inputs
Genomic DNA sequence tokens
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Results/Pre-training (paragraph 2); Results/GenomeOcean Learns Protein-coding Principles from DNA Alone (paragraph 5)Outputs
Generated DNA sequences and representations usable for task-specific adaptation
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Results/Model Safety (paragraph 1); Methods/Model Evaluation/Generating and Evaluating Synthetic Metagenomes (paragraph 1)limited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
Release 2026-09-17-d277315f7d76 · 1 evaluation · 3 metric rows. Different protocols are not a single leaderboard.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| GenomeOcean: Natural vs artificial microbial genome sequence Configuration: GenomeOceanProtocol: Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence)Dataset: GenomeOcean natural/artificial sequence test 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. Author-reported evaluation · Evaluation metadata: needs review | ||
| 99.03 F1 Unit: % · Direction: higher | Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported | source checkedGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies; GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Table 2, GenomeOcean row, F1 column Source checking is not independent reproduction. |
| 99.03% Precision Unit: percent · Direction: higher | Uncertainty: unreported Scored: Not reported · Eligible: Not reported | source checkedGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Table 2:, row GenomeOcean, column Precision; XML row4 column2 Source checking is not independent reproduction. |
| 99.03% Recall Unit: percent · Direction: higher | Uncertainty: unreported Scored: Not reported · Eligible: Not reported | source checkedGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Table 2:, row GenomeOcean, column Recall; XML row4 column3 Source checking is not independent reproduction. |
How it works
How the evaluated method works
A generative language model uses byte-pair tokenisation and architectural optimisations for DNA sequence modelling. Co-assembled metagenomic contigs provide the pretraining corpus, rather than only reference genomes.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1)What was evaluated
The linked evaluation record identifies GenomeOcean: Natural vs artificial microbial genome sequence. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-b3-027Strengths and limitations
Strengths and considerations
- The metagenomic corpus broadens representation beyond well-catalogued reference organisms.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Abstract (paragraph 1); Introduction (paragraph 4)
Limitations and conditions
- The paper’s sequence-generation and classification evaluations do not show that every generated sequence has an experimentally verified function.
Sources
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Results/GenomeOcean Learns Protein-coding Principles from DNA Alone (paragraph 6); Results/Inference Efficiency Benchmarking (paragraph 2)
Profile review details
Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction.
Stable record: reported-model-2df975e60d16d1Specifications
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 |
|---|---|
| Model type | Study-specific predictive method; this record is the paper-specific evaluated configuration.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) |
| Architecture / procedure | A generative language model uses byte-pair tokenisation and architectural optimisations for DNA sequence modelling. Co-assembled metagenomic contigs provide the pretraining corpus, rather than only reference genomes.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) |
| Biological inputs | Genomic DNA sequence tokensSourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Results/Pre-training (paragraph 2); Results/GenomeOcean Learns Protein-coding Principles from DNA Alone (paragraph 5) |
| Outputs | Generated DNA sequences and representations usable for task-specific adaptationSourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Results/Model Safety (paragraph 1); Methods/Model Evaluation/Generating and Evaluating Synthetic Metagenomes (paragraph 1) |
| Parameters | 4 billion parametersSourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Discussion (paragraph 5); Methods/Model Evaluation/Generating and Evaluating Synthetic Metagenomes (paragraph 2) |
| Known versions / configuration | GenomeOcean is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sourcesSourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | More than 600 Gbp of high-quality contigs assembled from 220 TB of metagenomic data.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Methods/Training/Evaluation Datasets/Assembled Metagenome Datasets (paragraph 1); Results/Pre-training (paragraph 1) |
| Context limits | Initial pretraining uses 1,024 BPE tokens. The 4B model is then continued at 10,240 tokens, approximately 51 kb; this larger limit is not assigned to the 100M or 500M variants.SourcesGenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Results/Pre-training (paragraph 1); Table T3 (paragraph 1) |
| Access | Official study implementation and usage documentation: https://github.com/jgi-genomeocean/genomeocean/blob/06fa433169539a3c84d7366a663933b888a5386b/README.md. This pinned documentation revision is not automatically the evaluated weight revision.Sourcesjgi-genomeocean/genomeocean README.md · README.md; installation, model download and usage instructions |
| Code licence | Custom Lawrence Berkeley/Northwestern permissive licence; see the pinned licence text for its conditions. (study repository code at the cited revision; this does not establish every dependency or historical checkpoint licence).Sourcesjgi-genomeocean/genomeocean LICENSE · LICENSE; complete licence text |
| Weights licence | The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. · Not reported in inspected sourcesSourcesjgi-genomeocean/genomeocean README.md · README.md; checkpoint/access documentation and licence scope |
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.
19 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram steps ["Genomic DNA sequence tokens","GenomeOcean","Generated DNA sequences and representations usable for task-specific adaptation"] Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Evaluated procedure (conceptual) Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Model type Study-specific predictive method; this record is the paper-specific evaluated configuration. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Architecture / procedure A generative language model uses byte-pair tokenisation and architectural optimisations for DNA sequence modelling. Co-assembled metagenomic contigs provide the pretraining corpus, rather than only reference genomes. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Weights licence The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. Individual claims | jgi-genomeocean/genomeocean README.md README.md; checkpoint/access documentation and licence scope Version: 06fa433169539a3c84d7366a663933b888a5386b | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Biological inputs Genomic DNA sequence tokens Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Results/Pre-training (paragraph 2); Results/GenomeOcean Learns Protein-coding Principles from DNA Alone (paragraph 5) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Outputs Generated DNA sequences and representations usable for task-specific adaptation Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Results/Model Safety (paragraph 1); Methods/Model Evaluation/Generating and Evaluating Synthetic Metagenomes (paragraph 1) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Parameters 4 billion parameters Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Discussion (paragraph 5); Methods/Model Evaluation/Generating and Evaluating Synthetic Metagenomes (paragraph 2) Version: preprint archived 2025-02-05 | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Known versions / configuration GenomeOcean is the comparison-table label; that label does not specify an immutable weight revision. Individual claims | GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. Version: preprint archived 2025-02-05 | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
3 source records and release history
- GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies · Original source · preprint archived 2025-02-05
- jgi-genomeocean/genomeocean README.md · Original source · 06fa433169539a3c84d7366a663933b888a5386b
- jgi-genomeocean/genomeocean LICENSE · Original source · 06fa433169539a3c84d7366a663933b888a5386b
Technical metadata and extraction receipts
Stable ID: reported-model-2df975e60d16d1
- areas
- microbes-communities
- entity level
- method
- version
- Not reported
- reported name
- GenomeOcean
- historical missing metadata
- version: not_reported_in_legacy_extract; checkpoint revision: not_reported_in_legacy_extract; training data: not_reported_in_legacy_extract; licence: 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
- model
- entity classification
- review date: 2026-09-17; rationale: This source-scoped entry preserves the method/configuration actually named in an evaluation. It is neither a global family identity nor proof of an immutable checkpoint; the linked evaluation retains adaptation, fitting and scoring details.; source ids: genomeocean-2025; source locator: Methods/Training/Evaluation Datasets/Evaluation of Dataset Complexity (paragraph 1); Abstract (paragraph 1) | Introduction (paragraph 5); Abstract (paragraph 1); ambiguities: Configuration means the source-labelled evaluated identity. It does not establish missing checkpoint hashes, default settings or equivalence to same-named records in other papers.