Contested result

DNABERT-2 and HyenaDNA: Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants reverses the order

Low severity
Not yet reviewed
Ordering reversal
Different tasks across studies

Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants ranks HyenaDNA above DNABERT-2 in 3 of 4 comparisons; other studies rank DNABERT-2 above HyenaDNA in 89 of 108.

An automatic lead, not a finding of error. The results below are shown as published, everywhere on the site, until a reviewer decides otherwise.

Which studies agree

3 papers rank DNABERT-2 above HyenaDNA in 89 of their 115 comparisons and HyenaDNA higher in 19; 7 are ties. Comparative Analysis of Deep Learning… ranks HyenaDNA above DNABERT-2 in 3 of its 4 comparisons and DNABERT-2 higher in 1.

Rank DNABERT-2 higher (3): GenBench, GENEB leaderboard, Nucleotide Transformer
Rank HyenaDNA higher (1): Comparative Analysis of Deep Learning… (disagrees)
Mixed (0): none

Each paper sits where at least two thirds of its comparisons point; Mixed means neither model reaches two thirds. The larger, outlined dot is the paper that disagrees.

DNABERT-2 higherTieHyenaDNA higherNumbers are counts of comparisons; the outlined bar is the paper that disagrees.

Papers compared (4)
Paper	Role	Ranks DNABERT-2 higher	Ranks HyenaDNA higher	Ties
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants · DOI	Disagrees	1	3	0
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models · Source	Agrees	43	4	4
GENEB leaderboard (leaderboard.json, 13 task-type aggregates) · Repository	Agrees	29	13	3
Nucleotide Transformer: building and evaluating robust foundation models for human genomics (Nature Methods) and 1 more record · DOI	Agrees	17	2	0

Comparisons in Comparative Analysis of Deep Learning… (4)

Protocol	Dataset	Metric	DNABERT-2	HyenaDNA	Higher
Allelic reporter effect correlation in HeLa (Manzo et al. 2025 Table 1)	HeLa regulatory variant reporter data (Manzo et al. 2025)	Pearson (predicted versus measured log2 fold-change of variant effect), unitless	0.12	0.055	DNABERT-2
Allelic reporter effect correlation in HepG2 (Manzo et al. 2025 Table 1)	HepG2 regulatory variant reporter data (Manzo et al. 2025)	Pearson (predicted versus measured log2 fold-change of variant effect), unitless	0.096	0.142	HyenaDNA
Allelic reporter effect correlation in K562 (Manzo et al. 2025 Table 1)	K562 regulatory variant reporter data (Manzo et al. 2025)	Pearson (predicted versus measured log2 fold-change of variant effect), unitless	0.086	0.148	HyenaDNA
Allelic reporter effect correlation in NPC (Manzo et al. 2025 Table 1)	NPC regulatory variant reporter data (Manzo et al. 2025)	Pearson (predicted versus measured log2 fold-change of variant effect), unitless	-0.004	0.002	HyenaDNA

Comparisons in GenBench (51)

Protocol	Dataset	Metric	DNABERT-2	HyenaDNA	Higher
GenBench Bulk RNA expression	GenBench Bulk RNA expression	MSE, unitless	0.451	0.453	Tie
GenBench Bulk RNA expression	GenBench Bulk RNA expression	Spearman, unitless	0.767	0.755	DNABERT-2
GenBench Drosophila enhancer activity (dev)	GenBench Drosophila enhancer activity (dev)	Pearson, unitless	0.617	0.47	DNABERT-2
GenBench Drosophila enhancer activity (hk)	GenBench Drosophila enhancer activity (hk)	Pearson, unitless	0.734	0.552	DNABERT-2
GenBench Drosophila enhancer activity (Mean)	GenBench Drosophila enhancer activity (Mean)	Pearson (mean of dev and hk as printed), unitless	0.678	0.511	DNABERT-2
GenBench Long-range Promoters Prediction	GenBench Long-range Promoters Prediction	Accuracy, fraction	0.976	0.887	DNABERT-2
GenBench Long-range Species Classification	GenBench Long-range Species Classification	Accuracy, fraction	0.747	0.677	DNABERT-2
GenBench Long-range splice site prediction (Mean)	GenBench Long-range splice site prediction (Mean)	AUROC (mean of donor and acceptor as printed), unitless	0.671	0.648	DNABERT-2
GenBench Long-range splice site prediction (Splice acceptor)	GenBench Long-range splice site prediction (Splice acceptor)	AUROC, unitless	0.707	0.723	HyenaDNA
GenBench Long-range splice site prediction (Splice donar)	GenBench Long-range splice site prediction (Splice donar)	AUROC, unitless	0.635	0.574	DNABERT-2
GenBench Coding vs Intergenomic	GenBench Coding vs Intergenomic	Accuracy (task-level value as printed in Table 4), fraction	0.936	0.91	DNABERT-2
GenBench Human Core Promoter Detection	GenBench Human Core Promoter Detection	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows), fraction	0.826	0.844	HyenaDNA
GenBench Human Core Promoter Detection, dataset all	GenBench Human Core Promoter Detection, dataset all	Accuracy, fraction	0.824	0.831	Tie
GenBench Human Core Promoter Detection, dataset notata	GenBench Human Core Promoter Detection, dataset notata	Accuracy, fraction	0.844	0.835	DNABERT-2
GenBench Human Core Promoter Detection, dataset tata	GenBench Human Core Promoter Detection, dataset tata	Accuracy, fraction	0.81	0.867	HyenaDNA
GenBench Human Enhancers Cohn	GenBench Human Enhancers Cohn	Accuracy (task-level value as printed in Table 4), fraction	0.759	0.73	DNABERT-2
GenBench Human Enhancers Ensembl	GenBench Human Enhancers Ensembl	Accuracy (task-level value as printed in Table 4), fraction	0.907	0.903	Tie
GenBench Human Ensembl Regulatory	GenBench Human Ensembl Regulatory	Accuracy (task-level value as printed in Table 4), fraction	0.883	0.846	DNABERT-2
GenBench Human Nontata Promoters	GenBench Human Nontata Promoters	Accuracy (task-level value as printed in Table 4), fraction	0.952	0.945	Tie
GenBench Human OCR Ensembl	GenBench Human OCR Ensembl	Accuracy (task-level value as printed in Table 4), fraction	0.758	0.791	HyenaDNA
GenBench Human Promoter Detection	GenBench Human Promoter Detection	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows), fraction	0.899	0.73	DNABERT-2
GenBench Human Promoter Detection, dataset all	GenBench Human Promoter Detection, dataset all	Accuracy, fraction	0.925	0.775	DNABERT-2
GenBench Human Promoter Detection, dataset notata	GenBench Human Promoter Detection, dataset notata	Accuracy, fraction	0.966	0.81	DNABERT-2
GenBench Human Promoter Detection, dataset tata	GenBench Human Promoter Detection, dataset tata	Accuracy, fraction	0.806	0.604	DNABERT-2
GenBench Human Splice Site Detection	GenBench Human Splice Site Detection	Accuracy (task-level value as printed in Table 4), fraction	0.881	0.566	DNABERT-2
GenBench Human Transcription Factor Prediction	GenBench Human Transcription Factor Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows), fraction	0.822	0.698	DNABERT-2
GenBench Human Transcription Factor Prediction, dataset 0	GenBench Human Transcription Factor Prediction, dataset 0	Accuracy, fraction	0.838	0.722	DNABERT-2
GenBench Human Transcription Factor Prediction, dataset 1	GenBench Human Transcription Factor Prediction, dataset 1	Accuracy, fraction	0.852	0.74	DNABERT-2
GenBench Human Transcription Factor Prediction, dataset 2	GenBench Human Transcription Factor Prediction, dataset 2	Accuracy, fraction	0.802	0.669	DNABERT-2
GenBench Human Transcription Factor Prediction, dataset 3	GenBench Human Transcription Factor Prediction, dataset 3	Accuracy, fraction	0.755	0.643	DNABERT-2
GenBench Human Transcription Factor Prediction, dataset 4	GenBench Human Transcription Factor Prediction, dataset 4	Accuracy, fraction	0.862	0.714	DNABERT-2
GenBench Human vs Worm	GenBench Human vs Worm	Accuracy (task-level value as printed in Table 4), fraction	0.974	0.962	DNABERT-2
GenBench Mouse Enhancers	GenBench Mouse Enhancers	Accuracy (task-level value as printed in Table 4), fraction	0.818	0.793	DNABERT-2
GenBench Mouse Transcription Factor Prediction	GenBench Mouse Transcription Factor Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows), fraction	0.827	0.653	DNABERT-2
GenBench Mouse Transcription Factor Prediction, dataset 0	GenBench Mouse Transcription Factor Prediction, dataset 0	Accuracy, fraction	0.744	0.616	DNABERT-2
GenBench Mouse Transcription Factor Prediction, dataset 1	GenBench Mouse Transcription Factor Prediction, dataset 1	Accuracy, fraction	0.919	0.728	DNABERT-2
GenBench Mouse Transcription Factor Prediction, dataset 2	GenBench Mouse Transcription Factor Prediction, dataset 2	Accuracy, fraction	0.897	0.706	DNABERT-2
GenBench Mouse Transcription Factor Prediction, dataset 3	GenBench Mouse Transcription Factor Prediction, dataset 3	Accuracy, fraction	0.842	0.621	DNABERT-2
GenBench Mouse Transcription Factor Prediction, dataset 4	GenBench Mouse Transcription Factor Prediction, dataset 4	Accuracy, fraction	0.733	0.597	DNABERT-2
GenBench Virus Covid Variant Classification	GenBench Virus Covid Variant Classification	Accuracy (task-level value as printed in Table 4), fraction	0.72	0.377	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction	GenBench Yeast Epigenetic Marks Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows), fraction	0.802	0.63	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3	GenBench Yeast Epigenetic Marks Prediction, dataset H3	Accuracy, fraction	0.897	0.685	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3K14ac	GenBench Yeast Epigenetic Marks Prediction, dataset H3K14ac	Accuracy, fraction	0.779	0.623	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3K36me3	GenBench Yeast Epigenetic Marks Prediction, dataset H3K36me3	Accuracy, fraction	0.797	0.633	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me1	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me1	Accuracy, fraction	0.726	0.594	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me2	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me2	Accuracy, fraction	0.734	0.609	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me3	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me3	Accuracy, fraction	0.76	0.569	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3K79me3	GenBench Yeast Epigenetic Marks Prediction, dataset H3K79me3	Accuracy, fraction	0.841	0.659	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H3K9ac	GenBench Yeast Epigenetic Marks Prediction, dataset H3K9ac	Accuracy, fraction	0.814	0.636	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H4	GenBench Yeast Epigenetic Marks Prediction, dataset H4	Accuracy, fraction	0.906	0.69	DNABERT-2
GenBench Yeast Epigenetic Marks Prediction, dataset H4ac	GenBench Yeast Epigenetic Marks Prediction, dataset H4ac	Accuracy, fraction	0.769	0.603	DNABERT-2

Comparisons in GENEB leaderboard (45)

Protocol	Dataset	Metric	DNABERT-2	HyenaDNA	Higher
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Chromatin Acc.'), unitless	0.32	0.34	HyenaDNA
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Coding/NC'), unitless	0.582	0.58	Tie
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'DNA Methyl.'), unitless	0.0414	0.0614	HyenaDNA
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Enhancers'), unitless	0.332	0.264	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Histone Mod.'), unitless	0.265	0.198	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Mouse Enh.'), unitless	0.0982	0.0903	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Promoters'), unitless	0.471	0.462	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Regulatory'), unitless	0.216	0.147	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Species Clf.'), unitless	0.402	0.385	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Splice Sites'), unitless	0.218	0.126	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'TF Binding'), unitless	0.147	0.158	HyenaDNA
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'Virus/Phage'), unitless	0.134	0.101	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; category 'lncRNA'), unitless	0.0627	0.0278	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; macro average: unweighted mean of the 13 category MCCs), unitless	0.253	0.222	DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregates	GENEB 100 tasks	MCC (10-shot; micro average: mean over all 100 tasks), unitless	0.272	0.23	DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Chromatin Acc.'), unitless	0.156	0.129	DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Coding/NC'), unitless	0.162	0.311	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'DNA Methyl.'), unitless	0.00287	0.0137	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Enhancers'), unitless	0.134	0.135	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Histone Mod.'), unitless	0.112	0.0859	DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Mouse Enh.'), unitless	0.0188	0.0258	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Promoters'), unitless	0.24	0.272	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Regulatory'), unitless	0.0709	0.0598	DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Species Clf.'), unitless	0.177	0.18	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Splice Sites'), unitless	0.0789	0.0644	DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'TF Binding'), unitless	0.0663	0.036	DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'Virus/Phage'), unitless	0.0421	0.0421	Tie
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; category 'lncRNA'), unitless	0.0163	0.0213	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; macro average: unweighted mean of the 13 category MCCs), unitless	0.0982	0.104	HyenaDNA
GENEB leaderboard, 1-shot, MCC aggregates	GENEB 100 tasks	MCC (1-shot; micro average: mean over all 100 tasks), unitless	0.119	0.117	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Chromatin Acc.'), unitless	0.57	0.562	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Coding/NC'), unitless	0.82	0.757	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'DNA Methyl.'), unitless	0.18	0.213	HyenaDNA
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Enhancers'), unitless	0.483	0.436	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Histone Mod.'), unitless	0.469	0.448	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Mouse Enh.'), unitless	0.527	0.468	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Promoters'), unitless	0.708	0.697	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Regulatory'), unitless	0.435	0.432	Tie
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Species Clf.'), unitless	0.625	0.666	HyenaDNA
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Splice Sites'), unitless	0.491	0.401	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'TF Binding'), unitless	0.574	0.563	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'Virus/Phage'), unitless	0.492	0.346	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; category 'lncRNA'), unitless	0.265	0.158	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; macro average: unweighted mean of the 13 category MCCs), unitless	0.511	0.467	DNABERT-2
GENEB leaderboard, full training data, MCC aggregates	GENEB 100 tasks	MCC (full training data; micro average: mean over all 100 tasks), unitless	0.506	0.472	DNABERT-2

Comparisons in Nucleotide Transformer (19)

Protocol	Dataset	Metric	DNABERT-2	HyenaDNA	Higher
NT benchmark Enhancer, fine-tuning (PEFT)	NT benchmark Enhancer	MCC (median over 10 cross-validation folds), unitless	0.517	0.476	DNABERT-2
NT benchmark Enhancer types, fine-tuning (PEFT)	NT benchmark Enhancer types	MCC (median over 10 cross-validation folds), unitless	0.476	0.445	DNABERT-2
NT benchmark H2AFZ, fine-tuning (PEFT)	NT benchmark H2AFZ	MCC (median over 10 cross-validation folds), unitless	0.49	0.467	DNABERT-2
NT benchmark H3K27ac, fine-tuning (PEFT)	NT benchmark H3K27ac	MCC (median over 10 cross-validation folds), unitless	0.491	0.423	DNABERT-2
NT benchmark H3K27me3, fine-tuning (PEFT)	NT benchmark H3K27me3	MCC (median over 10 cross-validation folds), unitless	0.599	0.55	DNABERT-2
NT benchmark H3K36me3, fine-tuning (PEFT)	NT benchmark H3K36me3	MCC (median over 10 cross-validation folds), unitless	0.637	0.553	DNABERT-2
NT benchmark H3K4me1, fine-tuning (PEFT)	NT benchmark H3K4me1	MCC (median over 10 cross-validation folds), unitless	0.49	0.43	DNABERT-2
NT benchmark H3K4me2, fine-tuning (PEFT)	NT benchmark H3K4me2	MCC (median over 10 cross-validation folds), unitless	0.558	0.521	DNABERT-2
NT benchmark H3K4me3, fine-tuning (PEFT)	NT benchmark H3K4me3	MCC (median over 10 cross-validation folds), unitless	0.646	0.603	DNABERT-2
NT benchmark H3K9ac, fine-tuning (PEFT)	NT benchmark H3K9ac	MCC (median over 10 cross-validation folds), unitless	0.564	0.487	DNABERT-2
NT benchmark H3K9me3, fine-tuning (PEFT)	NT benchmark H3K9me3	MCC (median over 10 cross-validation folds), unitless	0.443	0.419	DNABERT-2
NT benchmark H4K20me1, fine-tuning (PEFT)	NT benchmark H4K20me1	MCC (median over 10 cross-validation folds), unitless	0.655	0.59	DNABERT-2
NT benchmark normalised summed performance (Supplementary Table 9)	NT benchmark, all 18 tasks	MCC (mean over three task categories of median cross-validation MCC), unitless	0.723	0.691	DNABERT-2
NT benchmark Promoter all, fine-tuning (PEFT)	NT benchmark Promoter all	MCC (median over 10 cross-validation folds), unitless	0.754	0.698	DNABERT-2
NT benchmark Promoter non-TATA, fine-tuning (PEFT)	NT benchmark Promoter non-TATA	MCC (median over 10 cross-validation folds), unitless	0.769	0.729	DNABERT-2
NT benchmark Promoter TATA, fine-tuning (PEFT)	NT benchmark Promoter TATA	MCC (median over 10 cross-validation folds), unitless	0.784	0.666	DNABERT-2
NT benchmark Splice acceptor, fine-tuning (PEFT)	NT benchmark Splice acceptor	MCC (median over 10 cross-validation folds), unitless	0.837	0.815	DNABERT-2
NT benchmark Splice donor, fine-tuning (PEFT)	NT benchmark Splice donor	MCC (median over 10 cross-validation folds), unitless	0.861	0.943	HyenaDNA
NT benchmark Splice site all, fine-tuning (PEFT)	NT benchmark Splice site all	MCC (median over 10 cross-validation folds), unitless	0.855	0.907	HyenaDNA

What disagrees

Other studies: rank DNABERT-2 above HyenaDNA in 89 of 108 comparisons
This study: ranks HyenaDNA above DNABERT-2 in 3 of 4 comparisons

Compared: DNABERT-2, HyenaDNA
Study that disagrees: Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants
Other studies: Nucleotide Transformer: building and evaluating robust foundation models for human genomics (Nature Methods), GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models, GENEB leaderboard (leaderboard.json, 13 task-type aggregates), Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)
Rule: Ordering reversal; different tasks across studies
Model identity: Matched across studies partly through links not yet reviewed

The compared numbers

Each value links to its result record and source. Compare values within one protocol; the rule compares rankings, not raw values across protocols.

In the study that disagrees (8)
Study	Tool	Protocol or benchmark	Metric	Value
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	DNABERT-2 (Manzo et al. 2025)	Allelic reporter effect correlation in HeLa (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	0.12
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	Hyenadna 450 k (Manzo et al. 2025)	Allelic reporter effect correlation in HeLa (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	0.055
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	DNABERT-2 (Manzo et al. 2025)	Allelic reporter effect correlation in HepG2 (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	0.096
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	Hyenadna 160 k (Manzo et al. 2025)	Allelic reporter effect correlation in HepG2 (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	0.142
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	DNABERT-2 (Manzo et al. 2025)	Allelic reporter effect correlation in K562 (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	0.086
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	Hyenadna 160 k (Manzo et al. 2025)	Allelic reporter effect correlation in K562 (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	0.148
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	DNABERT-2 (Manzo et al. 2025)	Allelic reporter effect correlation in NPC (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	−0.004
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants	Hyenadna 32 k (Manzo et al. 2025)	Allelic reporter effect correlation in NPC (Manzo et al. 2025 Table 1)	Pearson (predicted versus measured log2 fold-change of variant effect)	0.002

In the other studies (216 results)

In the other studies
Study	Tool	Protocol or benchmark	Metric	Value
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Bulk RNA expression	Spearman	0.767
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Drosophila enhancer activity (dev)	Pearson	0.617
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Drosophila enhancer activity (hk)	Pearson	0.734
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Drosophila enhancer activity (Mean)	Pearson (mean of dev and hk as printed)	0.678
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Long-range Promoters Prediction	Accuracy	0.976
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Long-range Species Classification	Accuracy	0.747
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Long-range splice site prediction (Mean)	AUROC (mean of donor and acceptor as printed)	0.671
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Long-range splice site prediction (Splice acceptor)	AUROC	0.707
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Long-range splice site prediction (Splice donar)	AUROC	0.635
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Coding vs Intergenomic	Accuracy (task-level value as printed in Table 4)	0.936
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Core Promoter Detection	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.826
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Core Promoter Detection, dataset notata	Accuracy	0.844
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Core Promoter Detection, dataset tata	Accuracy	0.81
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Enhancers Cohn	Accuracy (task-level value as printed in Table 4)	0.759
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Ensembl Regulatory	Accuracy (task-level value as printed in Table 4)	0.883
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human OCR Ensembl	Accuracy (task-level value as printed in Table 4)	0.758
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Promoter Detection	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.899
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Promoter Detection, dataset all	Accuracy	0.925
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Promoter Detection, dataset notata	Accuracy	0.966
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Promoter Detection, dataset tata	Accuracy	0.806
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Splice Site Detection	Accuracy (task-level value as printed in Table 4)	0.881
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Transcription Factor Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.822
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Transcription Factor Prediction, dataset 0	Accuracy	0.838
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Transcription Factor Prediction, dataset 1	Accuracy	0.852
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Transcription Factor Prediction, dataset 2	Accuracy	0.802
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Transcription Factor Prediction, dataset 3	Accuracy	0.755
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human Transcription Factor Prediction, dataset 4	Accuracy	0.862
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Human vs Worm	Accuracy (task-level value as printed in Table 4)	0.974
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Mouse Enhancers	Accuracy (task-level value as printed in Table 4)	0.818
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Mouse Transcription Factor Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.827
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 0	Accuracy	0.744
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 1	Accuracy	0.919
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 2	Accuracy	0.897
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 3	Accuracy	0.842
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 4	Accuracy	0.733
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Virus Covid Variant Classification	Accuracy (task-level value as printed in Table 4)	0.72
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.802
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3	Accuracy	0.897
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K14ac	Accuracy	0.779
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K36me3	Accuracy	0.797
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me1	Accuracy	0.726
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me2	Accuracy	0.734
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me3	Accuracy	0.76
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K79me3	Accuracy	0.841
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K9ac	Accuracy	0.814
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H4	Accuracy	0.906
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	DNABERT2 (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H4ac	Accuracy	0.769
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Bulk RNA expression	Spearman	0.755
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Drosophila enhancer activity (dev)	Pearson	0.47
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Drosophila enhancer activity (hk)	Pearson	0.552
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Drosophila enhancer activity (Mean)	Pearson (mean of dev and hk as printed)	0.511
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Long-range Promoters Prediction	Accuracy	0.887
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Long-range Species Classification	Accuracy	0.677
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Long-range splice site prediction (Mean)	AUROC (mean of donor and acceptor as printed)	0.648
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Long-range splice site prediction (Splice acceptor)	AUROC	0.723
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Long-range splice site prediction (Splice donar)	AUROC	0.574
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Coding vs Intergenomic	Accuracy (task-level value as printed in Table 4)	0.91
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Core Promoter Detection	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.844
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Core Promoter Detection, dataset notata	Accuracy	0.835
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Core Promoter Detection, dataset tata	Accuracy	0.867
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Enhancers Cohn	Accuracy (task-level value as printed in Table 4)	0.73
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Ensembl Regulatory	Accuracy (task-level value as printed in Table 4)	0.846
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human OCR Ensembl	Accuracy (task-level value as printed in Table 4)	0.791
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Promoter Detection	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.73
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Promoter Detection, dataset all	Accuracy	0.775
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Promoter Detection, dataset notata	Accuracy	0.81
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Promoter Detection, dataset tata	Accuracy	0.604
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Splice Site Detection	Accuracy (task-level value as printed in Table 4)	0.566
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Transcription Factor Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.698
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Transcription Factor Prediction, dataset 0	Accuracy	0.722
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Transcription Factor Prediction, dataset 1	Accuracy	0.74
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Transcription Factor Prediction, dataset 2	Accuracy	0.669
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Transcription Factor Prediction, dataset 3	Accuracy	0.643
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human Transcription Factor Prediction, dataset 4	Accuracy	0.714
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Human vs Worm	Accuracy (task-level value as printed in Table 4)	0.962
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Mouse Enhancers	Accuracy (task-level value as printed in Table 4)	0.793
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Mouse Transcription Factor Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.653
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 0	Accuracy	0.616
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 1	Accuracy	0.728
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 2	Accuracy	0.706
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 3	Accuracy	0.621
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Mouse Transcription Factor Prediction, dataset 4	Accuracy	0.597
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Virus Covid Variant Classification	Accuracy (task-level value as printed in Table 4)	0.377
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction	Accuracy (task-level value as printed in Table 4; mean of the Table 10 dataset rows)	0.63
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3	Accuracy	0.685
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K14ac	Accuracy	0.623
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K36me3	Accuracy	0.633
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me1	Accuracy	0.594
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me2	Accuracy	0.609
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K4me3	Accuracy	0.569
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K79me3	Accuracy	0.659
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H3K9ac	Accuracy	0.636
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H4	Accuracy	0.69
GenBench: A Benchmarking Suite for Systematic Evaluation of Genomic Foundation Models	HyenaDNA (GenBench)	GenBench Yeast Epigenetic Marks Prediction, dataset H4ac	Accuracy	0.603
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Chromatin Acc.')	0.32
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'DNA Methyl.')	0.0414
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Enhancers')	0.332
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Histone Mod.')	0.265
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'lncRNA')	0.0627
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; macro average: unweighted mean of the 13 category MCCs)	0.253
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; micro average: mean over all 100 tasks)	0.272
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Mouse Enh.')	0.0982
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Promoters')	0.471
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Regulatory')	0.216
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Species Clf.')	0.402
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Splice Sites')	0.218
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'TF Binding')	0.147
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Virus/Phage')	0.134
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Chromatin Acc.')	0.156
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Coding/NC')	0.162
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'DNA Methyl.')	0.00287
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Enhancers')	0.134
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Histone Mod.')	0.112
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'lncRNA')	0.0163
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; macro average: unweighted mean of the 13 category MCCs)	0.0982
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; micro average: mean over all 100 tasks)	0.119
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Mouse Enh.')	0.0188
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Promoters')	0.24
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Regulatory')	0.0709
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Species Clf.')	0.177
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Splice Sites')	0.0789
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'TF Binding')	0.0663
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Chromatin Acc.')	0.57
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Coding/NC')	0.82
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'DNA Methyl.')	0.18
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Enhancers')	0.483
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Histone Mod.')	0.469
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'lncRNA')	0.265
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; macro average: unweighted mean of the 13 category MCCs)	0.511
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; micro average: mean over all 100 tasks)	0.506
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Mouse Enh.')	0.527
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Promoters')	0.708
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Species Clf.')	0.625
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Splice Sites')	0.491
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'TF Binding')	0.574
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	dnabert2 (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Virus/Phage')	0.492
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Chromatin Acc.')	0.34
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Enhancers')	0.264
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Histone Mod.')	0.198
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'lncRNA')	0.0278
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Promoters')	0.462
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Species Clf.')	0.385
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Virus/Phage')	0.101
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Chromatin Acc.')	0.129
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Coding/NC')	0.311
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Histone Mod.')	0.0859
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'lncRNA')	0.0213
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; macro average: unweighted mean of the 13 category MCCs)	0.104
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; micro average: mean over all 100 tasks)	0.117
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Promoters')	0.272
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Regulatory')	0.0598
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Species Clf.')	0.18
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Splice Sites')	0.0644
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Enhancers')	0.436
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Histone Mod.')	0.448
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'lncRNA')	0.158
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Species Clf.')	0.666
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Splice Sites')	0.401
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-large-1m (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Virus/Phage')	0.346
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'DNA Methyl.')	0.0614
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; macro average: unweighted mean of the 13 category MCCs)	0.222
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; micro average: mean over all 100 tasks)	0.23
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Mouse Enh.')	0.0903
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Regulatory')	0.147
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'Splice Sites')	0.126
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 10-shot, MCC aggregates	MCC (10-shot; category 'TF Binding')	0.158
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'DNA Methyl.')	0.0137
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Enhancers')	0.135
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'Mouse Enh.')	0.0258
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, 1-shot, MCC aggregates	MCC (1-shot; category 'TF Binding')	0.036
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Chromatin Acc.')	0.562
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Coding/NC')	0.757
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'DNA Methyl.')	0.213
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; macro average: unweighted mean of the 13 category MCCs)	0.467
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; micro average: mean over all 100 tasks)	0.472
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Mouse Enh.')	0.468
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'Promoters')	0.697
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)	hyenadna-medium-160k (GENEB leaderboard)	GENEB leaderboard, full training data, MCC aggregates	MCC (full training data; category 'TF Binding')	0.563
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Enhancer, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.517 (± 0.011)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Enhancer types, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.476 (± 0.009)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H2AFZ, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.49 (± 0.013)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K27ac, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.491 (± 0.01)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K27me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.599 (± 0.01)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K36me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.637 (± 0.007)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K4me1, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.49 (± 0.008)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K4me2, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.558 (± 0.013)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K4me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.646 (± 0.008)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K9ac, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.564 (± 0.013)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K9me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.443 (± 0.025)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H4K20me1, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.655 (± 0.011)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Promoter all, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.754 (± 0.009)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Promoter non-TATA, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.769 (± 0.009)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Promoter TATA, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.784 (± 0.036)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Splice acceptor, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.837 (± 0.006)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Splice donor, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.861 (± 0.004)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Splice site all, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.855 (± 0.005)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	DNABERT-2, parameter-efficient fine-tuning (NT benchmark)	NT benchmark normalised summed performance (Supplementary Table 9)	MCC (mean over three task categories of median cross-validation MCC)	0.723
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-1KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K27ac, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.423 (± 0.017)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-1KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K4me1, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.43 (± 0.014)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-1KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K4me2, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.521 (± 0.024)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-1KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Splice acceptor, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.815 (± 0.049)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-1KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Splice donor, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.943 (± 0.024)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Enhancer, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.476 (± 0.021)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Enhancer types, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.445 (± 0.009)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H2AFZ, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.467 (± 0.012)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K27me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.55 (± 0.009)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K36me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.553 (± 0.011)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K4me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.603 (± 0.02)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K9ac, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.487 (± 0.025)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H3K9me3, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.419 (± 0.03)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark H4K20me1, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.59 (± 0.007)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Promoter all, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.698 (± 0.011)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Promoter non-TATA, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.729 (± 0.009)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Promoter TATA, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.666 (± 0.041)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark Splice site all, fine-tuning (PEFT)	MCC (median over 10 cross-validation folds)	0.907 (± 0.018)
Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12)	HyenaDNA-32KB, parameter-efficient fine-tuning (NT benchmark)	NT benchmark normalised summed performance (Supplementary Table 9)	MCC (mean over three task categories of median cross-validation MCC)	0.691

Follow-up

Current state: Not yet reviewed.

No one has reviewed this finding yet.

How to help resolve this

If you know why these studies disagree, for example a different population, metric or preprocessing step, or you have a corrected value, tell us through the contribution form. A reviewer checks every submission against its source before anything changes, and the step is added to the follow-up above.

Finding discrepancy-646fb5eb5fccc11f · All contested results · How findings are made and followed up