rewirebio.iobenchmarks
Contested result

DNABERT-2 and Nucleotide Transformer: DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species Genomes reverses the order

  • Low severity
  • Not yet reviewed
  • Ordering reversal
  • unknown

DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species Genomes ranks DNABERT-2 above Nucleotide Transformer in 24 of 26 comparisons; 7 of 8 other studies rank Nucleotide Transformer above DNABERT-2.

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

10 papers rank Nucleotide Transformer above DNABERT-2 in 102 of their 146 comparisons and DNABERT-2 higher in 31; 13 are ties. Of those papers, BEND, GenomeOcean 2025 rank DNABERT-2 higher on their own (5 of 7, 3 of 3). DNABERT-2 ranks DNABERT-2 above Nucleotide Transformer in 24 of its 28 comparisons and Nucleotide Transformer higher in 2; 2 are ties.

Papers grouped by which system they rank higher: Nucleotide Transformer or DNABERT-2Rank Nucleotide Transformer higher: GENEB leaderboard, DART-Eval, Genomic heterogeneity inflates…, Benchmarking pre-trained genomic…, mRNABench, BacBench LICENSE file at commit 64b7d45, Comparative Analysis of Deep Learning…. Rank DNABERT-2 higher: DNABERT-2 (disagrees), BEND, GenomeOcean 2025. Mixed: Nucleotide Transformer.Rank Nucleotide Transformer h…MixedRank DNABERT-2 higherGENEB leaderboardGENEB leaderboard (leaderboard.json, 13 task-type aggregates)DART-EvalDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNAGenomic heterogeneityinflates…Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Benchmarkingpre-trained genomic…Benchmarking pre-trained genomic language models for RNA sequence-related predictive applicationsmRNABenchmRNABench: A curated benchmark for mature mRNA property and function predictionBacBench LICENSE fileat commit 64b7d45BacBench LICENSE file at commit 64b7d45Comparative Analysisof Deep Learning…Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory VariantsNucleotideTransformerNucleotide Transformer: building and evaluating robust foundation models for human genomics (Nature Methods)DNABERT-2DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesBENDBEND: Benchmarking DNA Language Models on Biologically Meaningful TasksGenomeOcean 2025GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies
  • Rank Nucleotide Transformer higher (7): GENEB leaderboard, DART-Eval, Genomic heterogeneity inflates…, Benchmarking pre-trained genomic…, mRNABench, BacBench LICENSE file at commit 64b7d45, Comparative Analysis of Deep Learning…
  • Rank DNABERT-2 higher (3): DNABERT-2 (disagrees), BEND, GenomeOcean 2025
  • Mixed (1): Nucleotide Transformer

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.

Comparisons in which each paper scores Nucleotide Transformer or DNABERT-2 higherDNABERT-2 (disagrees): Nucleotide Transformer higher in 2, DNABERT-2 higher in 24, 2 ties. GENEB leaderboard: Nucleotide Transformer higher in 35, DNABERT-2 higher in 8, 2 ties. DART-Eval: Nucleotide Transformer higher in 14, DNABERT-2 higher in 3, 5 ties. Nucleotide Transformer: Nucleotide Transformer higher in 7, DNABERT-2 higher in 8, 4 ties. Genomic heterogeneity inflates…: Nucleotide Transformer higher in 12, DNABERT-2 higher in 1, 1 tie. Benchmarking pre-trained genomic…: Nucleotide Transformer higher in 12, DNABERT-2 higher in 0, 0 ties. mRNABench: Nucleotide Transformer higher in 8, DNABERT-2 higher in 1, 1 tie. BacBench LICENSE file at commit 64b7d45: Nucleotide Transformer higher in 9, DNABERT-2 higher in 1, 0 ties. BEND: Nucleotide Transformer higher in 2, DNABERT-2 higher in 5, 0 ties. Comparative Analysis of Deep Learning…: Nucleotide Transformer higher in 3, DNABERT-2 higher in 1, 0 ties. GenomeOcean 2025: Nucleotide Transformer higher in 0, DNABERT-2 higher in 3, 0 ties.Disagrees: DNABERT-2DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species Genomes2 Nucleotide Transformer higher2 tie24 DNABERT-2 higher24GENEB leaderboardGENEB leaderboard (leaderboard.json, 13 task-type aggregates)35 Nucleotide Transformer higher352 tie8 DNABERT-2 higher8DART-EvalDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA14 Nucleotide Transformer higher145 tie53 DNABERT-2 higher3Nucleotide TransformerNucleotide Transformer: building and evaluating robust foundation models for human genomics (Nature Methods)7 Nucleotide Transformer higher74 tie48 DNABERT-2 higher8Genomic heterogeneity inflates…Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)12 Nucleotide Transformer higher121 tie1 DNABERT-2 higherBenchmarking pre-trained genomic…Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications12 Nucleotide Transformer higher12mRNABenchmRNABench: A curated benchmark for mature mRNA property and function prediction8 Nucleotide Transformer higher81 tie11 DNABERT-2 higher1BacBench LICENSE file at commit 64b7d45BacBench LICENSE file at commit 64b7d459 Nucleotide Transformer higher91 DNABERT-2 higher1BENDBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks2 Nucleotide Transformer higher25 DNABERT-2 higher5Comparative Analysis of Deep Learning…Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants3 Nucleotide Transformer higher31 DNABERT-2 higher1GenomeOcean 2025GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies3 DNABERT-2 higher3
Nucleotide Transformer higherTieDNABERT-2 higherNumbers are counts of comparisons; the outlined bar is the paper that disagrees.
Comparisons in DNABERT-2 (28)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
GUE CORE-PROMOTER-DETECTION-ALL: Core promoter detection, dataset allGUE Core promoter detection, all (GUE split)MCC, unitless67.469.4DNABERT-2
GUE CORE-PROMOTER-DETECTION-NOTATA: Core promoter detection, dataset notataGUE Core promoter detection, notata (GUE split)MCC, unitless67.568Tie
GUE CORE-PROMOTER-DETECTION-TATA: Core promoter detection, dataset tataGUE Core promoter detection, tata (GUE split)MCC, unitless69.774.2DNABERT-2
GUE COVID-VARIANT-CLASSIFICATION-COVID: Covid variant classification, dataset CovidGUE Covid variant classification, Covid (GUE split)Macro F1, percent66.771DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3: Epigenetic marks prediction, dataset H3GUE Epigenetic marks prediction, H3 (GUE split)MCC, unitless74.678.3DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3K14AC: Epigenetic marks prediction, dataset H3K14acGUE Epigenetic marks prediction, H3K14ac (GUE split)MCC, unitless44.152.6DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3K36ME3: Epigenetic marks prediction, dataset H3K36me3GUE Epigenetic marks prediction, H3K36me3 (GUE split)MCC, unitless50.956.9DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3K4ME1: Epigenetic marks prediction, dataset H3K4me1GUE Epigenetic marks prediction, H3K4me1 (GUE split)MCC, unitless43.150.5DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3K4ME2: Epigenetic marks prediction, dataset H3K4me2GUE Epigenetic marks prediction, H3K4me2 (GUE split)MCC, unitless30.331.1DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3K4ME3: Epigenetic marks prediction, dataset H3K4me3GUE Epigenetic marks prediction, H3K4me3 (GUE split)MCC, unitless30.936.3DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3K79ME3: Epigenetic marks prediction, dataset H3K79me3GUE Epigenetic marks prediction, H3K79me3 (GUE split)MCC, unitless61.267.4DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H3K9AC: Epigenetic marks prediction, dataset H3K9acGUE Epigenetic marks prediction, H3K9ac (GUE split)MCC, unitless52.455.6DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H4: Epigenetic marks prediction, dataset H4GUE Epigenetic marks prediction, H4 (GUE split)MCC, unitless79.880.7DNABERT-2
GUE EPIGENETIC-MARKS-PREDICTION-H4AC: Epigenetic marks prediction, dataset H4acGUE Epigenetic marks prediction, H4ac (GUE split)MCC, unitless41.550.4DNABERT-2
GUE PROMOTER-DETECTION-ALL: Promoter detection, dataset allGUE Promoter detection, all (GUE split)MCC, unitless9186.8Nucleotide Transformer
GUE PROMOTER-DETECTION-NOTATA: Promoter detection, dataset notataGUE Promoter detection, notata (GUE split)MCC, unitless93.194.3DNABERT-2
GUE PROMOTER-DETECTION-TATA: Promoter detection, dataset tataGUE Promoter detection, tata (GUE split)MCC, unitless75.871.6Nucleotide Transformer
GUE SPLICE-SITE-PREDICTION-RECONSTRUCT: Splice site prediction, dataset ReconstructGUE Splice site prediction, Reconstruct (GUE split)MCC, unitless85.885Tie
GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-0: Transcription factor prediction (human), dataset 0GUE Transcription factor prediction (human), 0 (GUE split)MCC, unitless66.372DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-1: Transcription factor prediction (human), dataset 1GUE Transcription factor prediction (human), 1 (GUE split)MCC, unitless68.376.1DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-2: Transcription factor prediction (human), dataset 2GUE Transcription factor prediction (human), 2 (GUE split)MCC, unitless58.766.5DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-3: Transcription factor prediction (human), dataset 3GUE Transcription factor prediction (human), 3 (GUE split)MCC, unitless49.158.5DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-4: Transcription factor prediction (human), dataset 4GUE Transcription factor prediction (human), 4 (GUE split)MCC, unitless67.677.4DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-0: Transcription factor prediction (mouse), dataset 0GUE Transcription factor prediction (mouse), 0 (GUE split)MCC, unitless48.356.8DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-1: Transcription factor prediction (mouse), dataset 1GUE Transcription factor prediction (mouse), 1 (GUE split)MCC, unitless8084.8DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-2: Transcription factor prediction (mouse), dataset 2GUE Transcription factor prediction (mouse), 2 (GUE split)MCC, unitless70.179.3DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-3: Transcription factor prediction (mouse), dataset 3GUE Transcription factor prediction (mouse), 3 (GUE split)MCC, unitless42.366.5DNABERT-2
GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-4: Transcription factor prediction (mouse), dataset 4GUE Transcription factor prediction (mouse), 4 (GUE split)MCC, unitless43.452.7DNABERT-2
Comparisons in GENEB leaderboard (45)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Chromatin Acc.'), unitless0.410.32Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Coding/NC'), unitless0.6670.582Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'DNA Methyl.'), unitless0.05750.0414Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Enhancers'), unitless0.3140.332DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Histone Mod.'), unitless0.2580.265DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Mouse Enh.'), unitless0.2020.0982Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Promoters'), unitless0.5320.471Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Regulatory'), unitless0.1520.216DNABERT-2
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Species Clf.'), unitless0.5120.402Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Splice Sites'), unitless0.2420.218Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'TF Binding'), unitless0.2010.147Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'Virus/Phage'), unitless0.1610.134Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; category 'lncRNA'), unitless0.1030.0627Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; macro average: unweighted mean of the 13 category MCCs), unitless0.2930.253Nucleotide Transformer
GENEB leaderboard, 10-shot, MCC aggregatesGENEB 100 tasksMCC (10-shot; micro average: mean over all 100 tasks), unitless0.30.272Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Chromatin Acc.'), unitless0.1690.156Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Coding/NC'), unitless0.2510.162Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'DNA Methyl.'), unitless0.01790.00287Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Enhancers'), unitless0.1710.134Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Histone Mod.'), unitless0.09520.112DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Mouse Enh.'), unitless0.04810.0188Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Promoters'), unitless0.3090.24Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Regulatory'), unitless0.08370.0709Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Species Clf.'), unitless0.1920.177Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Splice Sites'), unitless0.07680.0789DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'TF Binding'), unitless0.03610.0663DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'Virus/Phage'), unitless0.01940.0421DNABERT-2
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; category 'lncRNA'), unitless0.04540.0163Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; macro average: unweighted mean of the 13 category MCCs), unitless0.1160.0982Nucleotide Transformer
GENEB leaderboard, 1-shot, MCC aggregatesGENEB 100 tasksMCC (1-shot; micro average: mean over all 100 tasks), unitless0.1360.119Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Chromatin Acc.'), unitless0.5540.57DNABERT-2
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Coding/NC'), unitless0.8640.82Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'DNA Methyl.'), unitless0.3560.18Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Enhancers'), unitless0.4870.483Tie
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Histone Mod.'), unitless0.4980.469Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Mouse Enh.'), unitless0.5470.527Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Promoters'), unitless0.7390.708Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Regulatory'), unitless0.4880.435Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Species Clf.'), unitless0.7520.625Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Splice Sites'), unitless0.6520.491Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'TF Binding'), unitless0.5730.574Tie
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'Virus/Phage'), unitless0.5640.492Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; category 'lncRNA'), unitless0.3720.265Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; macro average: unweighted mean of the 13 category MCCs), unitless0.5730.511Nucleotide Transformer
GENEB leaderboard, full training data, MCC aggregatesGENEB 100 tasksMCC (full training data; micro average: mean over all 100 tasks), unitless0.5610.506Nucleotide Transformer
Comparisons in DART-Eval (22)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
DART-Eval CA-AUROC-GM12878: Chromatin activity prediction, GM12878, positives against negativesENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.9380.916Nucleotide Transformer
DART-Eval CA-AUROC-H1ESC: Chromatin activity prediction, H1ESC, positives against negativesENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.9580.94Nucleotide Transformer
DART-Eval CA-AUROC-HEPG2: Chromatin activity prediction, HEPG2, positives against negativesENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.9220.893Nucleotide Transformer
DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negativesENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.9750.963Nucleotide Transformer
DART-Eval CA-AUROC-K562: Chromatin activity prediction, K562, positives against negativesENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.9410.917Nucleotide Transformer
DART-Eval CA-SPEARMAN-GM12878: Chromatin activity prediction, GM12878, positives onlyENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)Spearman, unitless0.5150.489Nucleotide Transformer
DART-Eval CA-SPEARMAN-H1ESC: Chromatin activity prediction, H1ESC, positives onlyENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)Spearman, unitless0.7370.717Nucleotide Transformer
DART-Eval CA-SPEARMAN-HEPG2: Chromatin activity prediction, HEPG2, positives onlyENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)Spearman, unitless0.5130.472Nucleotide Transformer
DART-Eval CA-SPEARMAN-IMR90: Chromatin activity prediction, IMR90, positives onlyENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)Spearman, unitless0.4890.47Nucleotide Transformer
DART-Eval CA-SPEARMAN-K562: Chromatin activity prediction, K562, positives onlyENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)Spearman, unitless0.5830.529Nucleotide Transformer
DART-Eval CTS-ACC: Cell-type-specific element classification, overall accuracyENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)Accuracy, fraction0.6320.65DNABERT-2
DART-Eval CTS-GM12878: Cell-type-specific element classification, GM12878ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.880.894DNABERT-2
DART-Eval CTS-H1ESC: Cell-type-specific element classification, H1ESCENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.9250.93Tie
DART-Eval CTS-HEPG2: Cell-type-specific element classification, HEPG2ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.8810.891DNABERT-2
DART-Eval CTS-IMR90: Cell-type-specific element classification, IMR90ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.920.922Tie
DART-Eval CTS-K562: Cell-type-specific element classification, K562ENCODE chromatin accessibility peaks in five cell lines (DART-Eval split)AUROC, unitless0.8670.871Tie
DART-Eval REI-ABS: Regulatory element identification, absolute accuracyENCODE cCREs against dinucleotide-shuffled backgrounds (DART-Eval split)Accuracy, fraction0.920.913Tie
DART-Eval REI-PAIR: Regulatory element identification, paired accuracyENCODE cCREs against dinucleotide-shuffled backgrounds (DART-Eval split)Paired accuracy, fraction0.9760.973Tie
DART-Eval VS-AFRICAN-AUROC: Variant scoring on Chromatin QTLs in African LCLs, AUROCChromatin QTLs in African LCLs (DART-Eval split)AUROC, unitless0.6230.616Nucleotide Transformer
DART-Eval VS-AFRICAN-PEARSON_R: Variant scoring on Chromatin QTLs in African LCLs, Pearson rChromatin QTLs in African LCLs (DART-Eval split)Pearson, unitless0.230.184Nucleotide Transformer
DART-Eval VS-YORUBAN-AUROC: Variant scoring on DNase QTLs in Yoruban LCLs, AUROCDNase QTLs in Yoruban LCLs (DART-Eval split)AUROC, unitless0.670.631Nucleotide Transformer
DART-Eval VS-YORUBAN-PEARSON_R: Variant scoring on DNase QTLs in Yoruban LCLs, Pearson rDNase QTLs in Yoruban LCLs (DART-Eval split)Pearson, unitless0.5070.473Nucleotide Transformer
Comparisons in Nucleotide Transformer (19)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
NT benchmark Enhancer, fine-tuning (PEFT)NT benchmark EnhancerMCC (median over 10 cross-validation folds), unitless0.5040.517DNABERT-2
NT benchmark Enhancer types, fine-tuning (PEFT)NT benchmark Enhancer typesMCC (median over 10 cross-validation folds), unitless0.4690.476DNABERT-2
NT benchmark H2AFZ, fine-tuning (PEFT)NT benchmark H2AFZMCC (median over 10 cross-validation folds), unitless0.4780.49DNABERT-2
NT benchmark H3K27ac, fine-tuning (PEFT)NT benchmark H3K27acMCC (median over 10 cross-validation folds), unitless0.4860.491DNABERT-2
NT benchmark H3K27me3, fine-tuning (PEFT)NT benchmark H3K27me3MCC (median over 10 cross-validation folds), unitless0.6030.599Tie
NT benchmark H3K36me3, fine-tuning (PEFT)NT benchmark H3K36me3MCC (median over 10 cross-validation folds), unitless0.6320.637Tie
NT benchmark H3K4me1, fine-tuning (PEFT)NT benchmark H3K4me1MCC (median over 10 cross-validation folds), unitless0.4910.49Tie
NT benchmark H3K4me2, fine-tuning (PEFT)NT benchmark H3K4me2MCC (median over 10 cross-validation folds), unitless0.5690.558Nucleotide Transformer
NT benchmark H3K4me3, fine-tuning (PEFT)NT benchmark H3K4me3MCC (median over 10 cross-validation folds), unitless0.6150.646DNABERT-2
NT benchmark H3K9ac, fine-tuning (PEFT)NT benchmark H3K9acMCC (median over 10 cross-validation folds), unitless0.5290.564DNABERT-2
NT benchmark H3K9me3, fine-tuning (PEFT)NT benchmark H3K9me3MCC (median over 10 cross-validation folds), unitless0.4830.443Nucleotide Transformer
NT benchmark H4K20me1, fine-tuning (PEFT)NT benchmark H4K20me1MCC (median over 10 cross-validation folds), unitless0.6590.655Tie
NT benchmark normalised summed performance (Supplementary Table 9)NT benchmark, all 18 tasksMCC (mean over three task categories of median cross-validation MCC), unitless0.7380.723Nucleotide Transformer
NT benchmark Promoter all, fine-tuning (PEFT)NT benchmark Promoter allMCC (median over 10 cross-validation folds), unitless0.7080.754DNABERT-2
NT benchmark Promoter non-TATA, fine-tuning (PEFT)NT benchmark Promoter non-TATAMCC (median over 10 cross-validation folds), unitless0.7580.769DNABERT-2
NT benchmark Promoter TATA, fine-tuning (PEFT)NT benchmark Promoter TATAMCC (median over 10 cross-validation folds), unitless0.8020.784Nucleotide Transformer
NT benchmark Splice acceptor, fine-tuning (PEFT)NT benchmark Splice acceptorMCC (median over 10 cross-validation folds), unitless0.9520.837Nucleotide Transformer
NT benchmark Splice donor, fine-tuning (PEFT)NT benchmark Splice donorMCC (median over 10 cross-validation folds), unitless0.9630.861Nucleotide Transformer
NT benchmark Splice site all, fine-tuning (PEFT)NT benchmark Splice site allMCC (median over 10 cross-validation folds), unitless0.9560.855Nucleotide Transformer
Comparisons in Genomic heterogeneity inflates… (14)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: All variants (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, All variants), unitless0.7930.576Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Intron (Non-splice) (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Intron (Non-splice)), unitless0.750.437Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Missense (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Missense), unitless0.5940.484Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Missense & Intron (Non-splice) (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Missense & Intron (Non-splice)), unitless0.6640.449Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Missense & 3'UTR (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Missense & 3'UTR), unitless0.5460.546Tie
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Noncoding (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Noncoding), unitless0.8510.433Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: RNA gene (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, RNA gene), unitless0.3710.423DNABERT-2
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Splice (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Splice), unitless0.6170.454Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Start loss (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Start loss), unitless0.5690.389Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Stop gain (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Stop gain), unitless0.6180.483Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Stop loss (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Stop loss), unitless0.6090.57Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: Synonymous (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, Synonymous), unitless0.770.552Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: 3' UTR (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, 3' UTR), unitless0.5640.499Nucleotide Transformer
ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)Lu et al. 2025 ClinVar benchmark: 5' UTR (Figure 1)AUROC (ClinVar pathogenic vs benign SNVs, 5' UTR), unitless0.5670.414Nucleotide Transformer
Comparisons in Benchmarking pre-trained genomic… (12)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, acceptorsAccuracy (NT-splice-sites-acceptors, fine-tuned), fraction0.990.978Nucleotide Transformer
NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, acceptorsF1 (NT-splice-sites-acceptors, fine-tuned), fraction0.990.978Nucleotide Transformer
NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, acceptorsPrecision (NT-splice-sites-acceptors, fine-tuned), fraction0.990.978Nucleotide Transformer
NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, acceptorsRecall (NT-splice-sites-acceptors, fine-tuned), fraction0.990.978Nucleotide Transformer
NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, all sitesAccuracy (NT-splice-sites-all, fine-tuned), fraction0.9840.95Nucleotide Transformer
NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, all sitesF1 (NT-splice-sites-all, fine-tuned), fraction0.9840.95Nucleotide Transformer
NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, all sitesPrecision (NT-splice-sites-all, fine-tuned), fraction0.9840.95Nucleotide Transformer
NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, all sitesRecall (NT-splice-sites-all, fine-tuned), fraction0.9840.95Nucleotide Transformer
NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, donorsAccuracy (NT-splice-sites-donors, fine-tuned), fraction0.9910.968Nucleotide Transformer
NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, donorsF1 (NT-splice-sites-donors, fine-tuned), fraction0.9910.968Nucleotide Transformer
NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, donorsPrecision (NT-splice-sites-donors, fine-tuned), fraction0.9910.968Nucleotide Transformer
NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Nucleotide Transformer benchmark splice-site dataset, donorsRecall (NT-splice-sites-donors, fine-tuned), fraction0.9910.968Nucleotide Transformer
Comparisons in mRNABench (10)
Comparisons in BacBench LICENSE file at commit 64b7d45 (10)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
BacBench antibiotic resistance, binary linear probe AUPRCAntibiotic resistance binary prediction subset, 37 antibioticsAUPRC (linear head per drug on frozen genome embeddings, random 70/10/20 split, mean over 37 antibiotics), unitless59.752.4Nucleotide Transformer
BacBench antibiotic resistance, binary linear probe AUROCAntibiotic resistance binary prediction subset, 37 antibioticsAUROC (linear head per drug on frozen genome embeddings, random 70/10/20 split, mean over 37 antibiotics), unitless84.279.9Nucleotide Transformer
BacBench antibiotic resistance, MIC regression Pearson correlationAntibiotic resistance MIC regression subset, 56 antibioticsPearson (linear head per drug on frozen genome embeddings predicting log(1+MIC), random 70/10/20 split, mean over 56 antibiotics, printed as a percentage), unitless51.245.9Nucleotide Transformer
BacBench antibiotic resistance, MIC regression R squaredAntibiotic resistance MIC regression subset, 56 antibioticsR² (linear head per drug on frozen genome embeddings predicting log(1+MIC), random 70/10/20 split, mean over 56 antibiotics, printed as a percentage), unitless27.723.6Nucleotide Transformer
BacBench essential gene prediction, linear probe AUPRCBacBench essential gene dataset from the Database of Essential GenesAUPRC (linear head on frozen gene embeddings, genus-level split, median over held-out test genomes), unitless2630.2DNABERT-2
BacBench essential gene prediction, linear probe AUROCBacBench essential gene dataset from the Database of Essential GenesAUROC (linear head on frozen gene embeddings, genus-level split, median over held-out test genomes), unitless65.765Nucleotide Transformer
BacBench phenotypic trait prediction, linear probe AUPRCBacBench phenotypic trait datasetAUPRC (linear head per trait on frozen genome embeddings, macro-averaged across classes of each trait, genus-level split, mean over traits and seeds), unitless4341.9Nucleotide Transformer
BacBench phenotypic trait prediction, linear probe AUROCBacBench phenotypic trait datasetAUROC (linear head per trait on frozen genome embeddings, macro-averaged across classes of each trait, genus-level split, mean over traits and seeds), unitless6663.8Nucleotide Transformer
BacBench protein-protein interaction, linear probe AUPRCBacBench protein-protein interaction dataset from STRING DBAUPRC (linear head on frozen pair embeddings plus their cosine similarity, genus-level split, mean over held-out test genomes), unitless31.728.3Nucleotide Transformer
BacBench protein-protein interaction, linear probe AUROCBacBench protein-protein interaction dataset from STRING DBAUROC (linear head on frozen pair embeddings plus their cosine similarity, genus-level split, mean over held-out test genomes), unitless63.961.6Nucleotide Transformer
Comparisons in BEND (7)
Comparisons in Comparative Analysis of Deep Learning… (4)
ProtocolDatasetMetricNucleotide TransformerDNABERT-2Higher
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), unitless0.0870.12DNABERT-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), unitless0.1120.096Nucleotide Transformer
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), unitless0.1470.086Nucleotide Transformer
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), unitless0.005-0.004Nucleotide Transformer
Comparisons in GenomeOcean 2025 (3)

What disagrees

Other studies
rank Nucleotide Transformer above DNABERT-2 in 93 of 111 comparisons
This study
ranks DNABERT-2 above Nucleotide Transformer in 24 of 26 comparisons
Compared
DNABERT-2, Nucleotide Transformer
Study that disagrees
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species Genomes
Other studies
23 studiesBacBench LICENSE file at commit 64b7d45, BacBench repository README at commit 64b7d45, Bacformer LICENSE file at commit 61dca95, Bacformer repository README at commit 61dca95, Hugging Face model card for macwiatrak/bacformer-causal-complete-genomes, Hugging Face model card for macwiatrak/bacformer-causal-MAG, Hugging Face model card for macwiatrak/bacformer-for-ppi-prediction, Hugging Face model card for macwiatrak/bacformer-large-masked-complete-genomes, Hugging Face model card for macwiatrak/bacformer-large-masked-MAG, Hugging Face model card for macwiatrak/bacformer-masked-complete-genomes, Hugging Face model card for macwiatrak/bacformer-masked-MAG, Bacformer preprint, bioRxiv 2025.07.20.665723 version 2, Bacformer preprint, bioRxiv 2025.07.20.665723 version 3, DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA, mRNABench: A curated benchmark for mature mRNA property and function prediction, GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies, Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications, You et al. 2026, Supplementary Information, GENEB leaderboard (leaderboard.json, 13 task-type aggregates), Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2), mRNABench: A curated benchmark for mature mRNA property and function prediction, GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies, Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory Variants
Rule
Ordering reversal; unknown
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 (52)
StudyToolProtocol or benchmarkMetricValue
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE CORE-PROMOTER-DETECTION-ALL: Core promoter detection, dataset allMCC69.4
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE CORE-PROMOTER-DETECTION-TATA: Core promoter detection, dataset tataMCC74.2
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE COVID-VARIANT-CLASSIFICATION-COVID: Covid variant classification, dataset CovidMacro F171
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3: Epigenetic marks prediction, dataset H3MCC78.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3K14AC: Epigenetic marks prediction, dataset H3K14acMCC52.6
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3K36ME3: Epigenetic marks prediction, dataset H3K36me3MCC56.9
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3K4ME1: Epigenetic marks prediction, dataset H3K4me1MCC50.5
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3K4ME2: Epigenetic marks prediction, dataset H3K4me2MCC31.1
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3K4ME3: Epigenetic marks prediction, dataset H3K4me3MCC36.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3K79ME3: Epigenetic marks prediction, dataset H3K79me3MCC67.4
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H3K9AC: Epigenetic marks prediction, dataset H3K9acMCC55.6
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H4: Epigenetic marks prediction, dataset H4MCC80.7
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE EPIGENETIC-MARKS-PREDICTION-H4AC: Epigenetic marks prediction, dataset H4acMCC50.4
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE PROMOTER-DETECTION-ALL: Promoter detection, dataset allMCC86.8
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE PROMOTER-DETECTION-NOTATA: Promoter detection, dataset notataMCC94.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE PROMOTER-DETECTION-TATA: Promoter detection, dataset tataMCC71.6
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-0: Transcription factor prediction (human), dataset 0MCC72
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-1: Transcription factor prediction (human), dataset 1MCC76.1
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-2: Transcription factor prediction (human), dataset 2MCC66.5
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-3: Transcription factor prediction (human), dataset 3MCC58.5
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-4: Transcription factor prediction (human), dataset 4MCC77.4
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-0: Transcription factor prediction (mouse), dataset 0MCC56.8
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-1: Transcription factor prediction (mouse), dataset 1MCC84.8
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-2: Transcription factor prediction (mouse), dataset 2MCC79.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-3: Transcription factor prediction (mouse), dataset 3MCC66.5
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesDNABERT-2GUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-4: Transcription factor prediction (mouse), dataset 4MCC52.7
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE CORE-PROMOTER-DETECTION-ALL: Core promoter detection, dataset allMCC67.4
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE CORE-PROMOTER-DETECTION-TATA: Core promoter detection, dataset tataMCC69.7
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE COVID-VARIANT-CLASSIFICATION-COVID: Covid variant classification, dataset CovidMacro F166.7
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3: Epigenetic marks prediction, dataset H3MCC74.6
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3K14AC: Epigenetic marks prediction, dataset H3K14acMCC44.1
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3K36ME3: Epigenetic marks prediction, dataset H3K36me3MCC50.9
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3K4ME1: Epigenetic marks prediction, dataset H3K4me1MCC43.1
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3K4ME2: Epigenetic marks prediction, dataset H3K4me2MCC30.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3K4ME3: Epigenetic marks prediction, dataset H3K4me3MCC30.9
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3K79ME3: Epigenetic marks prediction, dataset H3K79me3MCC61.2
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H3K9AC: Epigenetic marks prediction, dataset H3K9acMCC52.4
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H4: Epigenetic marks prediction, dataset H4MCC79.8
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE EPIGENETIC-MARKS-PREDICTION-H4AC: Epigenetic marks prediction, dataset H4acMCC41.5
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE PROMOTER-DETECTION-ALL: Promoter detection, dataset allMCC91
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE PROMOTER-DETECTION-NOTATA: Promoter detection, dataset notataMCC93.1
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE PROMOTER-DETECTION-TATA: Promoter detection, dataset tataMCC75.8
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-0: Transcription factor prediction (human), dataset 0MCC66.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-1: Transcription factor prediction (human), dataset 1MCC68.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-2: Transcription factor prediction (human), dataset 2MCC58.7
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-3: Transcription factor prediction (human), dataset 3MCC49.1
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-HUMAN-4: Transcription factor prediction (human), dataset 4MCC67.6
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-0: Transcription factor prediction (mouse), dataset 0MCC48.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-1: Transcription factor prediction (mouse), dataset 1MCC80
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-2: Transcription factor prediction (mouse), dataset 2MCC70.1
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-3: Transcription factor prediction (mouse), dataset 3MCC42.3
DNABERT-2: Efficient Foundation Model and Benchmark for Multi-Species GenomesNT-2500M-1000gGUE TRANSCRIPTION-FACTOR-PREDICTION-MOUSE-4: Transcription factor prediction (mouse), dataset 4MCC43.4
In the other studies (222 results)
In the other studies
StudyToolProtocol or benchmarkMetricValue
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench antibiotic resistance, binary linear probe AUPRCAUPRC (linear head per drug on frozen genome embeddings, random 70/10/20 split, mean over 37 antibiotics)52.35 ± 23.55
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench antibiotic resistance, binary linear probe AUROCAUROC (linear head per drug on frozen genome embeddings, random 70/10/20 split, mean over 37 antibiotics)79.88 ± 7.26
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench antibiotic resistance, MIC regression Pearson correlationPearson (linear head per drug on frozen genome embeddings predicting log(1+MIC), random 70/10/20 split, mean over 56 antibiotics, printed as a percentage)45.89 ± 19.27
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench antibiotic resistance, MIC regression R squaredR² (linear head per drug on frozen genome embeddings predicting log(1+MIC), random 70/10/20 split, mean over 56 antibiotics, printed as a percentage)23.61 ± 18.33
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench antibiotic resistance, binary linear probe AUPRCAUPRC (linear head per drug on frozen genome embeddings, random 70/10/20 split, mean over 37 antibiotics)59.70 ± 22.90
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench antibiotic resistance, binary linear probe AUROCAUROC (linear head per drug on frozen genome embeddings, random 70/10/20 split, mean over 37 antibiotics)84.22 ± 6.47
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench antibiotic resistance, MIC regression Pearson correlationPearson (linear head per drug on frozen genome embeddings predicting log(1+MIC), random 70/10/20 split, mean over 56 antibiotics, printed as a percentage)51.19 ± 16.76
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench antibiotic resistance, MIC regression R squaredR² (linear head per drug on frozen genome embeddings predicting log(1+MIC), random 70/10/20 split, mean over 56 antibiotics, printed as a percentage)27.74 ± 17.80
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench essential gene prediction, linear probe AUPRCAUPRC (linear head on frozen gene embeddings, genus-level split, median over held-out test genomes)30.21 ± 19.81
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench essential gene prediction, linear probe AUROCAUROC (linear head on frozen gene embeddings, genus-level split, median over held-out test genomes)64.97 ± 8.29
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench essential gene prediction, linear probe AUPRCAUPRC (linear head on frozen gene embeddings, genus-level split, median over held-out test genomes)25.96 ± 21.26
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench essential gene prediction, linear probe AUROCAUROC (linear head on frozen gene embeddings, genus-level split, median over held-out test genomes)65.68 ± 8.02
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench phenotypic trait prediction, linear probe AUPRCAUPRC (linear head per trait on frozen genome embeddings, macro-averaged across classes of each trait, genus-level split, mean over traits and seeds)41.93 ± 34.31
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench phenotypic trait prediction, linear probe AUROCAUROC (linear head per trait on frozen genome embeddings, macro-averaged across classes of each trait, genus-level split, mean over traits and seeds)63.81 ± 11.15
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench phenotypic trait prediction, linear probe AUPRCAUPRC (linear head per trait on frozen genome embeddings, macro-averaged across classes of each trait, genus-level split, mean over traits and seeds)43.05 ± 34.91
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench phenotypic trait prediction, linear probe AUROCAUROC (linear head per trait on frozen genome embeddings, macro-averaged across classes of each trait, genus-level split, mean over traits and seeds)65.98 ± 11.54
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench protein-protein interaction, linear probe AUPRCAUPRC (linear head on frozen pair embeddings plus their cosine similarity, genus-level split, mean over held-out test genomes)28.30 ± 9.55
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3DNABERT-2 (frozen embeddings with a trained linear head)BacBench protein-protein interaction, linear probe AUROCAUROC (linear head on frozen pair embeddings plus their cosine similarity, genus-level split, mean over held-out test genomes)61.64 ± 3.58
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench protein-protein interaction, linear probe AUPRCAUPRC (linear head on frozen pair embeddings plus their cosine similarity, genus-level split, mean over held-out test genomes)31.66 ± 9.86
Bacformer preprint, bioRxiv 2025.07.20.665723 version 3Nucleotide Transformer (frozen embeddings with a trained linear head)BacBench protein-protein interaction, linear probe AUROCAUROC (linear head on frozen pair embeddings plus their cosine similarity, genus-level split, mean over held-out test genomes)63.92 ± 2.98
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-AUROC-GM12878: Chromatin activity prediction, GM12878, positives against negativesAUROC0.916
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-AUROC-H1ESC: Chromatin activity prediction, H1ESC, positives against negativesAUROC0.94
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-AUROC-HEPG2: Chromatin activity prediction, HEPG2, positives against negativesAUROC0.893
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negativesAUROC0.963
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-AUROC-K562: Chromatin activity prediction, K562, positives against negativesAUROC0.917
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-SPEARMAN-GM12878: Chromatin activity prediction, GM12878, positives onlySpearman0.489
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-SPEARMAN-H1ESC: Chromatin activity prediction, H1ESC, positives onlySpearman0.717
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-SPEARMAN-HEPG2: Chromatin activity prediction, HEPG2, positives onlySpearman0.472
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-SPEARMAN-IMR90: Chromatin activity prediction, IMR90, positives onlySpearman0.47
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CA-SPEARMAN-K562: Chromatin activity prediction, K562, positives onlySpearman0.529
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CTS-ACC: Cell-type-specific element classification, overall accuracyAccuracy0.65
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CTS-GM12878: Cell-type-specific element classification, GM12878AUROC0.894
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval CTS-HEPG2: Cell-type-specific element classification, HEPG2AUROC0.891
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval VS-AFRICAN-AUROC: Variant scoring on Chromatin QTLs in African LCLs, AUROCAUROC0.616
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval VS-AFRICAN-PEARSON_R: Variant scoring on Chromatin QTLs in African LCLs, Pearson rPearson0.184
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval VS-YORUBAN-AUROC: Variant scoring on DNase QTLs in Yoruban LCLs, AUROCAUROC0.631
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNADNABERT-2 (fine-tuned)DART-Eval VS-YORUBAN-PEARSON_R: Variant scoring on DNase QTLs in Yoruban LCLs, Pearson rPearson0.473
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-AUROC-GM12878: Chromatin activity prediction, GM12878, positives against negativesAUROC0.938
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-AUROC-H1ESC: Chromatin activity prediction, H1ESC, positives against negativesAUROC0.958
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-AUROC-HEPG2: Chromatin activity prediction, HEPG2, positives against negativesAUROC0.922
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-AUROC-IMR90: Chromatin activity prediction, IMR90, positives against negativesAUROC0.975
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-AUROC-K562: Chromatin activity prediction, K562, positives against negativesAUROC0.941
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-SPEARMAN-GM12878: Chromatin activity prediction, GM12878, positives onlySpearman0.515
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-SPEARMAN-H1ESC: Chromatin activity prediction, H1ESC, positives onlySpearman0.737
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-SPEARMAN-HEPG2: Chromatin activity prediction, HEPG2, positives onlySpearman0.513
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-SPEARMAN-IMR90: Chromatin activity prediction, IMR90, positives onlySpearman0.489
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CA-SPEARMAN-K562: Chromatin activity prediction, K562, positives onlySpearman0.583
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CTS-ACC: Cell-type-specific element classification, overall accuracyAccuracy0.632
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CTS-GM12878: Cell-type-specific element classification, GM12878AUROC0.88
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval CTS-HEPG2: Cell-type-specific element classification, HEPG2AUROC0.881
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval VS-AFRICAN-AUROC: Variant scoring on Chromatin QTLs in African LCLs, AUROCAUROC0.623
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval VS-AFRICAN-PEARSON_R: Variant scoring on Chromatin QTLs in African LCLs, Pearson rPearson0.23
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval VS-YORUBAN-AUROC: Variant scoring on DNase QTLs in Yoruban LCLs, AUROCAUROC0.67
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNANucleotide Transformer (fine-tuned)DART-Eval VS-YORUBAN-PEARSON_R: Variant scoring on DNase QTLs in Yoruban LCLs, Pearson rPearson0.507
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Accuracy (NT-splice-sites-acceptors, fine-tuned)0.9776(0.0005)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)F1 (NT-splice-sites-acceptors, fine-tuned)0.9776(0.0005)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Precision (NT-splice-sites-acceptors, fine-tuned)0.9777(0.0005)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Recall (NT-splice-sites-acceptors, fine-tuned)0.9776(0.0006)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Accuracy (NT-splice-sites-acceptors, fine-tuned)0.9899(0.0014)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)F1 (NT-splice-sites-acceptors, fine-tuned)0.9899(0.0014)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Precision (NT-splice-sites-acceptors, fine-tuned)0.9899(0.0014)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-acceptors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Recall (NT-splice-sites-acceptors, fine-tuned)0.9900(0.0014)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Accuracy (NT-splice-sites-all, fine-tuned)0.9499(0.0037)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)F1 (NT-splice-sites-all, fine-tuned)0.9500(0.0036)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Precision (NT-splice-sites-all, fine-tuned)0.9503(0.0034)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Recall (NT-splice-sites-all, fine-tuned)0.9499(0.0037)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Accuracy (NT-splice-sites-all, fine-tuned)0.9842(0.0009)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)F1 (NT-splice-sites-all, fine-tuned)0.9842(0.0009)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Precision (NT-splice-sites-all, fine-tuned)0.9843(0.0009)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-all splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Recall (NT-splice-sites-all, fine-tuned)0.9842(0.0009)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Accuracy (NT-splice-sites-donors, fine-tuned)0.9677(0.0029)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)F1 (NT-splice-sites-donors, fine-tuned)0.9677(0.0029)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Precision (NT-splice-sites-donors, fine-tuned)0.9678(0.0028)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationDNABERT2 fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Recall (NT-splice-sites-donors, fine-tuned)0.9677(0.0029)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Accuracy (NT-splice-sites-donors, fine-tuned)0.9913(0.0008)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)F1 (NT-splice-sites-donors, fine-tuned)0.9913(0.0008)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Precision (NT-splice-sites-donors, fine-tuned)0.9914(0.0008)
Benchmarking pre-trained genomic language models for RNA sequence-related predictive applications; You et al. 2026, Supplementary InformationNucleotideTransformer fine-tuned with an appended convolutional layer (You et al. 2026)NT-splice-sites-donors splice-site classification, fine-tuned models (You et al. 2026 Supplementary Table 4)Recall (NT-splice-sites-donors, fine-tuned)0.9913(0.0008)
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies; GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic AssembliesDNABERT-2Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence)F185.1
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Chromatin Acc.')0.32
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Coding/NC')0.582
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'DNA Methyl.')0.0414
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Enhancers')0.332
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Histone Mod.')0.265
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'lncRNA')0.0627
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (10-shot; category 'Mouse Enh.')0.0982
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Promoters')0.471
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Regulatory')0.216
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Species Clf.')0.402
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Splice Sites')0.218
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'TF Binding')0.147
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Virus/Phage')0.134
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Chromatin Acc.')0.156
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Coding/NC')0.162
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'DNA Methyl.')0.00287
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Enhancers')0.134
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Histone Mod.')0.112
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'lncRNA')0.0163
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (1-shot; category 'Mouse Enh.')0.0188
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Promoters')0.24
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Regulatory')0.0709
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Species Clf.')0.177
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Splice Sites')0.0789
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'TF Binding')0.0663
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Virus/Phage')0.0421
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (full training data; category 'lncRNA')0.265
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (full training data; category 'Promoters')0.708
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Regulatory')0.435
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)dnabert2 (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (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 aggregatesMCC (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 aggregatesMCC (full training data; category 'Virus/Phage')0.492
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Chromatin Acc.')0.41
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Coding/NC')0.667
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'DNA Methyl.')0.0575
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Enhancers')0.314
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Histone Mod.')0.258
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'lncRNA')0.103
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; macro average: unweighted mean of the 13 category MCCs)0.293
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; micro average: mean over all 100 tasks)0.3
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Mouse Enh.')0.202
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Promoters')0.532
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Regulatory')0.152
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Species Clf.')0.512
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Splice Sites')0.242
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'TF Binding')0.201
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 10-shot, MCC aggregatesMCC (10-shot; category 'Virus/Phage')0.161
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Chromatin Acc.')0.169
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Coding/NC')0.251
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'DNA Methyl.')0.0179
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Enhancers')0.171
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Histone Mod.')0.0952
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'lncRNA')0.0454
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; macro average: unweighted mean of the 13 category MCCs)0.116
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; micro average: mean over all 100 tasks)0.136
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Mouse Enh.')0.0481
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Promoters')0.309
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Regulatory')0.0837
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Species Clf.')0.192
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Splice Sites')0.0768
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'TF Binding')0.0361
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, 1-shot, MCC aggregatesMCC (1-shot; category 'Virus/Phage')0.0194
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Chromatin Acc.')0.554
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Coding/NC')0.864
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'DNA Methyl.')0.356
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Histone Mod.')0.498
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'lncRNA')0.372
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; macro average: unweighted mean of the 13 category MCCs)0.573
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; micro average: mean over all 100 tasks)0.561
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Mouse Enh.')0.547
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Promoters')0.739
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Regulatory')0.488
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Species Clf.')0.752
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Splice Sites')0.652
GENEB leaderboard (leaderboard.json, 13 task-type aggregates)NT-2.5B-MS (GENEB leaderboard)GENEB leaderboard, full training data, MCC aggregatesMCC (full training data; category 'Virus/Phage')0.564
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, All variants)0.576
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, All variants)0.793
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Intron (Non-splice))0.437
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Intron (Non-splice))0.75
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Missense & Intron (Non-splice))0.449
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Missense & Intron (Non-splice))0.664
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Missense)0.484
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Missense)0.594
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Noncoding)0.433
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Noncoding)0.851
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, RNA gene)0.423
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, RNA gene)0.371
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Splice)0.454
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Splice)0.617
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Start loss)0.389
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Start loss)0.569
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Stop gain)0.483
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Stop gain)0.618
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Stop loss)0.57
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Stop loss)0.609
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Synonymous)0.552
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, Synonymous)0.77
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, 3' UTR)0.499
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, 3' UTR)0.564
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)DNABERT-2 117M, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, 5' UTR)0.414
Genomic heterogeneity inflates the performance of variant pathogenicity predictions (Lu, Liu, Lin and Brandes, bioRxiv 2025, v2)Nucleotide Transformer 2.5B multi-species, CLS embedding L2 distance (Lu et al. 2025)ClinVar pathogenic versus benign SNVs within variant type, AUROC (Lu et al. 2025)AUROC (ClinVar pathogenic vs benign SNVs, 5' UTR)0.567
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench ECLIP: eCLIP binding site predictionAUPRC37.1
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench GO: Gene Ontology term predictionAUPRC32.6
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench HL: mRNA half lifePearson0.48
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench MRL-HL-PAIR: Paired mean ribosome load and half lifePearson0.53
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench MRL-MPRA: Mean ribosome load on an MPRA libraryPearson0.57
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench MRL: Mean ribosome loadPearson0.29
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench MRNA-LOC-SR: mRNA localisation, short rangeAUPRC67.4
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench PROT-LOC: Protein localisationAUPRC32.5
mRNABench: A curated benchmark for mature mRNA property and function predictionDNABERT2mRNABench VEP: Variant effect predictionAUPRC26.4
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench ECLIP: eCLIP binding site predictionAUPRC40.5
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench GO: Gene Ontology term predictionAUPRC36
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench HL: mRNA half lifePearson0.57
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench MRL-HL-PAIR: Paired mean ribosome load and half lifePearson0.57
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench MRL-MPRA: Mean ribosome load on an MPRA libraryPearson0.6
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench MRL: Mean ribosome loadPearson0.34
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench MRNA-LOC-SR: mRNA localisation, short rangeAUPRC72.2
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench PROT-LOC: Protein localisationAUPRC32
mRNABench: A curated benchmark for mature mRNA property and function predictionNTmRNABench VEP: Variant effect predictionAUPRC31.8
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic AssembliesNucleotide Transformers V2Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence)F183.1
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies; GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic AssembliesDNABERT-2Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence)Recall85
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic AssembliesNucleotide Transformers V2Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence)Recall83
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic Assemblies; GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic AssembliesDNABERT-2Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence)Precision85.2
GenomeOcean: An Efficient Genome Foundation Model Trained on Large-Scale Metagenomic AssembliesNucleotide Transformers V2Natural versus GenomeOcean-generated DNA classification (Natural vs artificial microbial genome sequence)Precision83.3
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory VariantsDNABERT-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 VariantsNT 2.5b-1000g (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)0.087
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory VariantsDNABERT-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 VariantsNT 2.5b-1000g (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)0.112
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory VariantsDNABERT-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 VariantsNT 2.5b-1000g (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)0.147
Comparative Analysis of Deep Learning Models for Predicting Causative Regulatory VariantsDNABERT-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 VariantsNT 2.5b-1000g (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)0.005

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-38fc71356acdd111 · All contested results · How findings are made and followed up