DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
Chromatin activity prediction, IMR90, positives against negatives. Scored with AUROC on ENCODE chromatin accessibility peaks in five cell lines. Separating positive IMR90 peaks from matched negatives.
Overview
Chromatin activity prediction, IMR90, positives against negatives. Scored with AUROC on ENCODE chromatin accessibility peaks in five cell lines. Separating positive IMR90 peaks from matched negatives.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
Benchmarks
These source-backed links do not make different protocols or scores interchangeable.
Recorded evaluations
Each evaluation records what was tested and under which conditions.
- Caduceus (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- Caduceus (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- ChromBPNet (ab initio) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- DNABERT-2 (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- GENA-LM (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- GENA-LM (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- HyenaDNA (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- HyenaDNA (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- Mistral-DNA (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- Mistral-DNA (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- Nucleotide Transformer (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
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.
Published comparisons
Explore the results reported under one evaluation protocol. Each figure keeps its source, dataset and metric together; it is not a ranking across studies. The pooled view gathers every source table that reports the same metric and names what it does not hold constant.
DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
auroc (fraction) · Higher values are better for this metric.
Every method DART-Eval reports on Chromatin activity prediction, IMR90, positives against negatives, scored with AUROC on ENCODE chromatin accessibility peaks in five cell lines.
Evaluation protocol · ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)
- Caduceus (probed) · Configuration · Author-reported evaluation0.610
- DNABERT-2 (probed) · Configuration · Author-reported evaluation0.729
- GENA-LM (probed) · Configuration · Author-reported evaluation0.799
- HyenaDNA (probed) · Configuration · Author-reported evaluation0.702
- Mistral-DNA (probed) · Configuration · Author-reported evaluation0.712
- Nucleotide Transformer (probed) · Configuration · Author-reported evaluation0.739
- Caduceus (fine-tuned) · Configuration · Author-reported evaluation0.976
- DNABERT-2 (fine-tuned) · Configuration · Author-reported evaluation0.963
- GENA-LM (fine-tuned) · Configuration · Author-reported evaluation0.961
- HyenaDNA (fine-tuned) · Configuration · Author-reported evaluation0.941
- Mistral-DNA (fine-tuned) · Configuration · Author-reported evaluation0.855
- Nucleotide Transformer (fine-tuned) · Configuration · Author-reported evaluation0.975
- ChromBPNet (ab initio) · Method · Author-reported evaluation0.975
Source order is preserved. Plotted marks show point estimates; uncertainty, where reported, is retained in the printed values and table. Differences do not establish statistical significance.
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, column(AUROC IMR90)Values, uncertainty and evidence
| Tested entity | Printed value | Uncertainty | Evidence |
|---|---|---|---|
| Caduceus (probed) · Configuration | 0.610 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Probed Caduceus), column(CA-AUROC-IMR90) |
| DNABERT-2 (probed) · Configuration | 0.729 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Probed DNABERT-2), column(CA-AUROC-IMR90) |
| GENA-LM (probed) · Configuration | 0.799 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Probed GENA-LM), column(CA-AUROC-IMR90) |
| HyenaDNA (probed) · Configuration | 0.702 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Probed HyenaDNA), column(CA-AUROC-IMR90) |
| Mistral-DNA (probed) · Configuration | 0.712 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Probed Mistral-DNA), column(CA-AUROC-IMR90) |
| Nucleotide Transformer (probed) · Configuration | 0.739 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Probed NT), column(CA-AUROC-IMR90) |
| Caduceus (fine-tuned) · Configuration | 0.976 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned Caduceus), column(CA-AUROC-IMR90) |
| DNABERT-2 (fine-tuned) · Configuration | 0.963 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-AUROC-IMR90) |
| GENA-LM (fine-tuned) · Configuration | 0.961 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned GENA-LM), column(CA-AUROC-IMR90) |
| HyenaDNA (fine-tuned) · Configuration | 0.941 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned HyenaDNA), column(CA-AUROC-IMR90) |
| Mistral-DNA (fine-tuned) · Configuration | 0.855 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned Mistral-DNA), column(CA-AUROC-IMR90) |
| Nucleotide Transformer (fine-tuned) · Configuration | 0.975 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned NT), column(CA-AUROC-IMR90) |
| ChromBPNet (ab initio) · Method | 0.975 fraction | Not reported | Author-reported evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Ab initio ChromBPNet), column(CA-AUROC-IMR90) |
Scope and limitations
- Author-reported numbers, source checked but not independently reproduced.
- The evaluation setting is part of the method name: a zero-shot, probed and fine-tuned run of the same model are different entries.
- Metrics and datasets differ between tasks, so these figures cannot be averaged into one score.
Source transcription and grouping reviewed by automated source review on 2026-09-18. These experiments were not independently reproduced by rewire.
Tested entities and results
Release 2026-09-17-134cd1815de8 · 13 evaluations · 13 metric rows. Different protocols are not a single leaderboard. Where several source tables report the same metric, the published comparisons above offer a pooled view that names what it does not hold constant.
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.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: part of discovery-benchmark-dart-eval Individual claims | DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA Table 5, column(AUROC IMR90) Version: 2412.05430v1 | source checked automated source review · 2026-09-18 Audit detailsPrimary-source transcription with no human sign-off and no independent reproduction. Field: Claim: dart-eval-association-ca-auroc-imr90 Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
Sources and history
Release 2026-09-17-134cd1815de8 · Record review: source checked
1 source records and release history
- DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Original source · 2412.05430v1
Technical metadata and extraction receipts
Stable ID: dart-eval-task-ca-auroc-imr90
- areas
- dna-genomes
- tasks
- Chromatin activity prediction, IMR90, positives against negatives
- metric
- AUROC
- metric direction
- higher
- dataset
- ENCODE chromatin accessibility peaks in five cell lines
- protocol
- Separating positive IMR90 peaks from matched negatives.
- source locator
- Table 5, column(AUROC IMR90)
- comparison panels
- id: dart-eval-panel-ca-auroc-imr90; title: DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives; protocol id: dart-eval-task-ca-auroc-imr90; dataset id: dart-eval-dataset-encode-chromatin-accessibility-peaks-in-five-cell-lines; metric: auroc; unit: fraction; direction: higher; result ids: dart-eval-result-caduceus-probed-ca-auroc-imr90-auroc; dart-eval-result-dnabert-2-probed-ca-auroc-imr90-auroc; dart-eval-result-gena-lm-probed-ca-auroc-imr90-auroc; dart-eval-result-hyenadna-probed-ca-auroc-imr90-auroc; dart-eval-result-mistral-dna-probed-ca-auroc-imr90-auroc; dart-eval-result-nucleotide-transformer-probed-ca-auroc-imr90-auroc; dart-eval-result-caduceus-fine-tuned-ca-auroc-imr90-auroc; dart-eval-result-dnabert-2-fine-tuned-ca-auroc-imr90-auroc; dart-eval-result-gena-lm-fine-tuned-ca-auroc-imr90-auroc; dart-eval-result-hyenadna-fine-tuned-ca-auroc-imr90-auroc; dart-eval-result-mistral-dna-fine-tuned-ca-auroc-imr90-auroc; dart-eval-result-nucleotide-transformer-fine-tuned-ca-auroc-imr90-auroc; dart-eval-result-chrombpnet-ab-initio-ca-auroc-imr90-auroc; source ids: evidence-expansion-p2-evidence-discovery-final-dart-4194b137ba55; source locator: Table 5, column(AUROC IMR90); context: Every method DART-Eval reports on Chromatin activity prediction, IMR90, positives against negatives, scored with AUROC on ENCODE chromatin accessibility peaks in five cell lines.; caveats: Author-reported numbers, source checked but not independently reproduced.; The evaluation setting is part of the method name: a zero-shot, probed and fine-tuned run of the same model are different entries.; Metrics and datasets differ between tasks, so these figures cannot be averaged into one score.; review: method: automated_source_review; date: 2026-09-18
Related records
- part of: DART-Eval
- subject: DART-Eval CA-AUROC-IMR90: part of discovery-benchmark-dart-eval
- benchmark: Caduceus (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: Caduceus (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: ChromBPNet (ab initio) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: DNABERT-2 (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: GENA-LM (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: GENA-LM (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: HyenaDNA (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: HyenaDNA (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: Mistral-DNA (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: Mistral-DNA (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: Nucleotide Transformer (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- benchmark: Nucleotide Transformer (probed) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives