Model type
DNA sequence transformer; this record is the paper-specific evaluated configuration.
Configuration
DNABERT-2
DART-Eval tests DNABERT-2 on regulatory DNA under separately defined zero-shot, probed and fine-tuned protocols.
Sources
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation proceduresHow 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
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation proceduresOverview
Biological inputs
Regulatory DNA and matched controls defined by the DART-Eval task
Sources
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation proceduresOutputs
Regulatory-element scores or task-specific predictions, according to the evaluation protocol
Sources
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation proceduresAccess
Official upstream implementation and usage documentation: https://github.com/MAGICS-LAB/DNABERT_2/blob/f25bed9ee20db966dff39e5c1571249d04e36404/README.md. This pinned documentation revision is not automatically the evaluated weight revision.
Sources
MAGICS-LAB/DNABERT_2 README.md · README.md; installation, model download and usage instructionslimited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
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 |
|---|---|---|---|
| Configuration: DNABERT-2 | Task: regulatory element identification Dataset: DART-Eval cCREs versus matched shuffled controls | 0.876 accuracy fraction · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourceDNABERT-2: regulatory element identification zero-shot likelihood ranking: higher likelihood for cCRE than matched control Aggregation: Not reported DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 3 (PDF page 5), DNABERT-2 row, Zero-Shot Accuracy column |
Source checking is not independent reproduction. Release 2026-09-29-06401fd5b220.
Use this model
How it works, versions and access
Related profile: DNABERT-2. This page retains the exact record and its evaluation context.
How it works
How the evaluated method works
The masked DNA transformer uses byte-pair tokenisation. The linked regulatory-element row uses the paper’s zero-shot procedure; trained probing or fine-tuning rows are distinct evaluations.
Sources
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation proceduresUnderlying method and version boundaries
DNABERT-2 replaces overlapping k-mer tokens with byte-pair encoding and uses ALiBi positional biases. The official 117M model produces 768-dimensional token representations; downstream classifiers and pooling choices are separate configuration details.
Sources
MAGICS-LAB/DNABERT_2 README.md · README.md; introduction, model description, pretrained-model and usage sections at pinned revisionWhat was evaluated
The linked evaluation record identifies DNABERT-2: regulatory element identification. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-b2-dart-eval-regulatory-2024Strengths, limitations and unresolved questions
Strengths and limitations
Strengths and considerations
- Separates zero-shot representation evidence from the effect of training a downstream predictor.
Sources
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures
Limitations and conditions
- Regulatory-element discrimination does not establish quantitative cell-specific accessibility or regulatory-variant accuracy.
Sources
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures
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-28413ae1766316Specifications
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 | DNA sequence transformer; this record is the paper-specific evaluated configuration.SourcesMAGICS-LAB/DNABERT_2 README.md · README.md model description |
| Architecture / procedure | The masked DNA transformer uses byte-pair tokenisation. The linked regulatory-element row uses the paper’s zero-shot procedure; trained probing or fine-tuning rows are distinct evaluations.SourcesDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures |
| Biological inputs | Regulatory DNA and matched controls defined by the DART-Eval taskSourcesDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures |
| Outputs | Regulatory-element scores or task-specific predictions, according to the evaluation protocolSourcesDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures |
| Parameters | 117 million parametersSourcesDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures |
| Known versions / configuration | DNABERT-2 is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sourcesSourcesDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | The benchmark identifies multispecies genomic pretraining; task adaptation is reported separately.SourcesDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures |
| Context limits | A maximum input/context length for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sourcesSources (2)DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA; MAGICS-LAB/DNABERT_2 README.md · Complete primary text and named comparison table; inspected for explicit maximum input length (dataset lengths and family-wide limits are not substituted); README.md at pinned repository revision |
| Access | Official upstream implementation and usage documentation: https://github.com/MAGICS-LAB/DNABERT_2/blob/f25bed9ee20db966dff39e5c1571249d04e36404/README.md. This pinned documentation revision is not automatically the evaluated weight revision.SourcesMAGICS-LAB/DNABERT_2 README.md · README.md; installation, model download and usage instructions |
| Code licence | Apache 2.0 (upstream repository code at the cited revision; this does not establish every dependency or historical checkpoint licence).SourcesMAGICS-LAB/DNABERT_2 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 sourcesSourcesMAGICS-LAB/DNABERT_2 README.md · README.md; checkpoint/access documentation and licence scope |
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
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.
24 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 | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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
| DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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 | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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 DNA sequence transformer; this record is the paper-specific evaluated configuration. Individual claims | MAGICS-LAB/DNABERT_2 README.md README.md model description Version: f25bed9ee20db966dff39e5c1571249d04e36404 | 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 The masked DNA transformer uses byte-pair tokenisation. The linked regulatory-element row uses the paper’s zero-shot procedure; trained probing or fine-tuning rows are distinct evaluations. Individual claims | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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 | MAGICS-LAB/DNABERT_2 README.md README.md; checkpoint/access documentation and licence scope Version: f25bed9ee20db966dff39e5c1571249d04e36404 | 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 Regulatory DNA and matched controls defined by the DART-Eval task Individual claims | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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 Regulatory-element scores or task-specific predictions, according to the evaluation protocol Individual claims | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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 117 million parameters Individual claims | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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 DNABERT-2 is the comparison-table label; that label does not specify an immutable weight revision. Individual claims | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. Version: NeurIPS 2024 Datasets and Benchmarks Track proceedings | 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
View linked audit checks and correction history
Release 2026-09-29-06401fd5b220 · Record review: needs review
3 source records and release history
- DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Original source · NeurIPS 2024 Datasets and Benchmarks Track proceedings
- MAGICS-LAB/DNABERT_2 README.md · Original source · f25bed9ee20db966dff39e5c1571249d04e36404
- MAGICS-LAB/DNABERT_2 LICENSE · Original source · f25bed9ee20db966dff39e5c1571249d04e36404
Technical metadata and extraction receipts
Stable ID: reported-model-28413ae1766316
- areas
- dna-genomes
- entity level
- method
- version
- not stated in table
- reported name
- DNABERT-2
- historical missing metadata
- 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: dart-eval-regulatory-2024; evidence-reported-base-dnabert2-readme-md; source locator: Table 2 (models); Section 3.2 zero-shot analysis; Table 3 regulatory-element identification; Appendix B evaluation procedures | README.md model description; 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.
Related records
- family: DNABERT-2
- model: DNABERT-2: regulatory element identification
- subject: DNABERT-2: family DNABERT-2