DNABERT-2 (fine-tuned)
DNA language model evaluated by the DART-Eval authors in the fine-tuned setting.
Overview
DNA language model evaluated by the DART-Eval authors in the fine-tuned setting.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluations and results
Release 2026-09-17-134cd1815de8 · 22 evaluations · 22 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.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-GM12878: Chromatin activity prediction, GM12878, positives against negatives Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-AUROC-GM12878: Chromatin activity prediction, GM12878, positives against negativesDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Separating positive GM12878 peaks from matched negatives. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.916 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-AUROC-GM12878) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-H1ESC: Chromatin activity prediction, H1ESC, positives against negatives Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-AUROC-H1ESC: Chromatin activity prediction, H1ESC, positives against negativesDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Separating positive H1ESC peaks from matched negatives. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.940 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-AUROC-H1ESC) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-HEPG2: Chromatin activity prediction, HEPG2, positives against negatives Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-AUROC-HEPG2: Chromatin activity prediction, HEPG2, positives against negativesDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Separating positive HEPG2 peaks from matched negatives. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.893 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-AUROC-HEPG2) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negativesDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Separating positive IMR90 peaks from matched negatives. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.963 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-AUROC-IMR90) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-K562: Chromatin activity prediction, K562, positives against negatives Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-AUROC-K562: Chromatin activity prediction, K562, positives against negativesDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Separating positive K562 peaks from matched negatives. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.917 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-AUROC-K562) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-GM12878: Chromatin activity prediction, GM12878, positives only Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-SPEARMAN-GM12878: Chromatin activity prediction, GM12878, positives onlyDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Rank correlation with measured accessibility among positive GM12878 peaks. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.489 spearman_r Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-SPEARMAN-GM12878) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-H1ESC: Chromatin activity prediction, H1ESC, positives only Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-SPEARMAN-H1ESC: Chromatin activity prediction, H1ESC, positives onlyDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Rank correlation with measured accessibility among positive H1ESC peaks. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.717 spearman_r Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-SPEARMAN-H1ESC) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-HEPG2: Chromatin activity prediction, HEPG2, positives only Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-SPEARMAN-HEPG2: Chromatin activity prediction, HEPG2, positives onlyDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Rank correlation with measured accessibility among positive HEPG2 peaks. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.472 spearman_r Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-SPEARMAN-HEPG2) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-IMR90: Chromatin activity prediction, IMR90, positives only Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-SPEARMAN-IMR90: Chromatin activity prediction, IMR90, positives onlyDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Rank correlation with measured accessibility among positive IMR90 peaks. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.470 spearman_r Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-SPEARMAN-IMR90) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-K562: Chromatin activity prediction, K562, positives only Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CA-SPEARMAN-K562: Chromatin activity prediction, K562, positives onlyDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Rank correlation with measured accessibility among positive K562 peaks. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.529 spearman_r Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 5, row(Fine-Tuned DNABERT-2), column(CA-SPEARMAN-K562) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CTS-ACC: Cell-type-specific element classification, overall accuracy Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CTS-ACC: Cell-type-specific element classification, overall accuracyDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) Classify which of five cell lines a accessible element belongs to. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.650 accuracy Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 4, row(Fine-Tuned DNABERT-2), column(CTS-ACC) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CTS-GM12878: Cell-type-specific element classification, GM12878 Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CTS-GM12878: Cell-type-specific element classification, GM12878Dataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) One-against-rest AUROC for GM12878 accessible elements. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.894 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 4, row(Fine-Tuned DNABERT-2), column(CTS-GM12878) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CTS-H1ESC: Cell-type-specific element classification, H1ESC Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CTS-H1ESC: Cell-type-specific element classification, H1ESCDataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) One-against-rest AUROC for H1ESC accessible elements. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.930 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 4, row(Fine-Tuned DNABERT-2), column(CTS-H1ESC) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CTS-HEPG2: Cell-type-specific element classification, HEPG2 Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CTS-HEPG2: Cell-type-specific element classification, HEPG2Dataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) One-against-rest AUROC for HEPG2 accessible elements. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.891 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 4, row(Fine-Tuned DNABERT-2), column(CTS-HEPG2) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CTS-IMR90: Cell-type-specific element classification, IMR90 Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CTS-IMR90: Cell-type-specific element classification, IMR90Dataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) One-against-rest AUROC for IMR90 accessible elements. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.922 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 4, row(Fine-Tuned DNABERT-2), column(CTS-IMR90) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval CTS-K562: Cell-type-specific element classification, K562 Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval CTS-K562: Cell-type-specific element classification, K562Dataset subset: ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split) One-against-rest AUROC for K562 accessible elements. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.871 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 4, row(Fine-Tuned DNABERT-2), column(CTS-K562) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval REI-ABS: Regulatory element identification, absolute accuracy Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval REI-ABS: Regulatory element identification, absolute accuracyDataset subset: ENCODE cCREs against dinucleotide-shuffled backgrounds (DART-Eval split) Distinguish ENCODE cCREs from dinucleotide-shuffled background sequences. The setting the model was run in is part of the method name. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.913 absolute_accuracy Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 3, row(DNABERT-2), column(fine-tuned absolute_accuracy) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval REI-PAIR: Regulatory element identification, paired accuracy Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval REI-PAIR: Regulatory element identification, paired accuracyDataset subset: ENCODE cCREs against dinucleotide-shuffled backgrounds (DART-Eval split) Distinguish ENCODE cCREs from dinucleotide-shuffled background sequences. The setting the model was run in is part of the method name. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.973 paired_accuracy Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 3, row(DNABERT-2), column(fine-tuned paired_accuracy) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval VS-AFRICAN-AUROC: Variant scoring on Chromatin QTLs in African LCLs, AUROC Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval VS-AFRICAN-AUROC: Variant scoring on Chromatin QTLs in African LCLs, AUROCDataset subset: Chromatin QTLs in African LCLs (DART-Eval split) Score the effect of a variant on chromatin accessibility, against the measured QTL call. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.616 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 6, row(African DNABERT-2), column(fine-tuned auroc) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval VS-AFRICAN-PEARSON_R: Variant scoring on Chromatin QTLs in African LCLs, Pearson r Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval VS-AFRICAN-PEARSON_R: Variant scoring on Chromatin QTLs in African LCLs, Pearson rDataset subset: Chromatin QTLs in African LCLs (DART-Eval split) Score the effect of a variant on chromatin accessibility, against the measured QTL call. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.184 pearson_r Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 6, row(African DNABERT-2), column(fine-tuned pearson_r) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval VS-YORUBAN-AUROC: Variant scoring on DNase QTLs in Yoruban LCLs, AUROC Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval VS-YORUBAN-AUROC: Variant scoring on DNase QTLs in Yoruban LCLs, AUROCDataset subset: DNase QTLs in Yoruban LCLs (DART-Eval split) Score the effect of a variant on chromatin accessibility, against the measured QTL call. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.631 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 6, row(Yoruban DNABERT-2), column(fine-tuned auroc) Source checking is not independent reproduction. |
| DNABERT-2 (fine-tuned) on DART-Eval VS-YORUBAN-PEARSON_R: Variant scoring on DNase QTLs in Yoruban LCLs, Pearson r Configuration: DNABERT-2 (fine-tuned)Task: DART-Eval VS-YORUBAN-PEARSON_R: Variant scoring on DNase QTLs in Yoruban LCLs, Pearson rDataset subset: DNase QTLs in Yoruban LCLs (DART-Eval split) Score the effect of a variant on chromatin accessibility, against the measured QTL call. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.473 pearson_r Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 6, row(Yoruban DNABERT-2), column(fine-tuned pearson_r) Source checking is not 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.
0 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
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-method-dnabert-2-fine-tuned
- areas
- dna-genomes
- source locator
- Table 1, row(DNABERT-2)
- missing metadata
- checkpoint revision: unreported; parameters: unextracted
Related records
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-GM12878: Chromatin activity prediction, GM12878, positives against negatives
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-H1ESC: Chromatin activity prediction, H1ESC, positives against negatives
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-HEPG2: Chromatin activity prediction, HEPG2, positives against negatives
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negatives
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-AUROC-K562: Chromatin activity prediction, K562, positives against negatives
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-GM12878: Chromatin activity prediction, GM12878, positives only
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-H1ESC: Chromatin activity prediction, H1ESC, positives only
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-HEPG2: Chromatin activity prediction, HEPG2, positives only
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-IMR90: Chromatin activity prediction, IMR90, positives only
- model: DNABERT-2 (fine-tuned) on DART-Eval CA-SPEARMAN-K562: Chromatin activity prediction, K562, positives only
- model: DNABERT-2 (fine-tuned) on DART-Eval CTS-ACC: Cell-type-specific element classification, overall accuracy
- model: DNABERT-2 (fine-tuned) on DART-Eval CTS-GM12878: Cell-type-specific element classification, GM12878
- model: DNABERT-2 (fine-tuned) on DART-Eval CTS-H1ESC: Cell-type-specific element classification, H1ESC
- model: DNABERT-2 (fine-tuned) on DART-Eval CTS-HEPG2: Cell-type-specific element classification, HEPG2
- model: DNABERT-2 (fine-tuned) on DART-Eval CTS-IMR90: Cell-type-specific element classification, IMR90
- model: DNABERT-2 (fine-tuned) on DART-Eval CTS-K562: Cell-type-specific element classification, K562
- model: DNABERT-2 (fine-tuned) on DART-Eval REI-ABS: Regulatory element identification, absolute accuracy
- model: DNABERT-2 (fine-tuned) on DART-Eval REI-PAIR: Regulatory element identification, paired accuracy
- model: DNABERT-2 (fine-tuned) on DART-Eval VS-AFRICAN-AUROC: Variant scoring on Chromatin QTLs in African LCLs, AUROC
- model: DNABERT-2 (fine-tuned) on DART-Eval VS-AFRICAN-PEARSON_R: Variant scoring on Chromatin QTLs in African LCLs, Pearson r
- model: DNABERT-2 (fine-tuned) on DART-Eval VS-YORUBAN-AUROC: Variant scoring on DNase QTLs in Yoruban LCLs, AUROC
- model: DNABERT-2 (fine-tuned) on DART-Eval VS-YORUBAN-PEARSON_R: Variant scoring on DNase QTLs in Yoruban LCLs, Pearson r