Task

Enhancer classification

Enhancer classification is evaluated within two published genomic sequence benchmark collections.

Sources

Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions

2 evaluations · 2 results

Overview

Datasets

Genomic Benchmarks includes human enhancer datasets; Nucleotide Transformer tasks include enhancer/non-enhancer and enhancer-strength labels.

Metrics

Classification accuracy is described for the benchmark comparison.

Allowed inputs

Genomic sequence.

Sources

Evaluation procedure diagram

How it worksComputational evaluation flow

Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol.

Sources (2)

Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision; enbed-2024__vbae117_supplementary_data.pdf · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; Supplement: Nucleotide Transformer benchmark dataset description and Table 1

limited source coverage · Automated source review, 2026-09-16. All specifications and missing details

Results

Results are available, but no reviewed comparison panel is linked in this release.

All evaluations

2 evaluations · 2 results. Different protocols are not a single leaderboard.

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: ENBED	Task: Enhancer classification Dataset: Genomic Benchmarks Mouse Enhancers	90.3 Accuracy % · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ENBED: Enhancer classification Reported Genomic Benchmarks classification accuracy. Aggregation: Not reported Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Table 2, Mouse Enhancers row, ENBED column
Configuration: ENBED (GRCh38)	Task: Enhancer classification Dataset: Genomic Benchmarks Mouse Enhancers	81.1 Accuracy % · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ENBED (GRCh38): Enhancer classification ENBED trained on GRCh38; reported Genomic Benchmarks classification accuracy. Aggregation: Not reported Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Table 2, Mouse Enhancers row, ENBED (GRCh38) column

Source checking is not independent reproduction. Release 2026-09-29-06401fd5b220.

Methods and evaluation design

Procedure, tasks and evaluated configurations

How it works

Evaluation methodology

Genomic Benchmarks includes human enhancer datasets; Nucleotide Transformer tasks include enhancer/non-enhancer and enhancer-strength labels. For the Nucleotide Transformer enhancer tasks, Supplementary Table 1 lists 14,968 training sequences and 400 test sequences of length 200. This set combines the original strong/weak/non-enhancer collection with 6,000 synthetic enhancers and 6,000 synthetic non-enhancers. Those counts do not describe the separate Genomic Benchmarks enhancer datasets. Classification accuracy is described for the benchmark comparison. Enformer, DNABERT-2, Nucleotide Transformer and HyenaDNA in the enhancer-containing NT task comparison. The enhancer-specific methods and supplement do not specify a chromosome holdout, homology threshold or synthetic-parent separation across these splits. The clade split and 95% similarity threshold described later in the supplement apply to mutation generation, not this enhancer evaluation.

Sources (2)

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Run instructions

No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths and considerations

No source-reviewed explanatory claims are recorded here yet.

Limitations and conditions

ENBED reports separate enhancer identification and enhancer-type tasks. Their splits and synthetic augmentation must stay distinct from the mutation-generation experiment’s phylogenetic holdouts.
Sources (2)
Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision; enbed-2024__vbae117_supplementary_data.pdf · Supplement: Nucleotide Transformer benchmark dataset description and Table 1; Supplement: Nucleotide Transformer benchmark description and Table 1; separate Mutation generation section

Profile review details

Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change.

Stable record: reported-task-132da895d4c381

Specifications

Inputs, training, access and other details

Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.

Data, procedure and scoring
Property	Description and evidence
Datasets	Genomic Benchmarks includes human enhancer datasets; Nucleotide Transformer tasks include enhancer/non-enhancer and enhancer-strength labels. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Splits	For the Nucleotide Transformer enhancer tasks, Supplementary Table 1 lists 14,968 training sequences and 400 test sequences of length 200. This set combines the original strong/weak/non-enhancer collection with 6,000 synthetic enhancers and 6,000 synthetic non-enhancers. Those counts do not describe the separate Genomic Benchmarks enhancer datasets. Sources (2) Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision; enbed-2024__vbae117_supplementary_data.pdf · Supplement: Nucleotide Transformer benchmark dataset description and Table 1
Metrics	Classification accuracy is described for the benchmark comparison. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Baselines	Enformer, DNABERT-2, Nucleotide Transformer and HyenaDNA in the enhancer-containing NT task comparison. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Leakage controls	The enhancer-specific methods and supplement do not specify a chromosome holdout, homology threshold or synthetic-parent separation across these splits. The clade split and 95% similarity threshold described later in the supplement apply to mutation generation, not this enhancer evaluation. · Not reported in inspected sources Sources (2) Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision; enbed-2024__vbae117_supplementary_data.pdf · Supplement: Nucleotide Transformer benchmark description and Table 1; separate Mutation generation section
Uncertainty	The cited text-accessible evaluation sections give no confidence-interval, resampling or repeat-run error-bar specification. Image-only tables and uninspected supplements are outside this absence claim. · Not reported in inspected sources Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Entity type	Paper-specific computational evaluation protocol. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Organisms	Human enhancer datasets within Genomic Benchmarks and Nucleotide Transformer tasks. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Assays	Enhancer identity/strength annotations. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Allowed inputs	Genomic sequence. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions
Adaptation	Task-level enhancer classification; exact task configuration must be selected from the evaluated benchmark collection. Sources Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Papers and result coverage

Last literature check: 2026-09-17. Dated primary-source discovery and protocol/table screening. Source checking does not mean experimental reproduction. Only separately extracted and independently reviewed numeric batches are publishable.

Paper or primary resource	Version	Reference
Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision	version of record	Read source

Historical gaps recorded on 2026-09-17

The catalogue now holds 2 result rows for this benchmark. A note below about pending extraction describes the state on 2026-09-17 and may since have been answered by a later batch. The result rows and their sources are the current record.

complete numerical transcription and independent cell review: Full primary artifact and table inventory preserved; no new numeric row is published from this audit alone.
exact checkpoint hashes and per-method scored denominators: Table labels alone do not establish these fields; do not infer checkpoint or scored count from model name or dataset size.

Search and extraction details

primary comparison tables located

Searches

Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision 10.1093/bioadv/vbae117

Evidence locations

Table 1.; XML table vbae117-T1
Table 2.; XML table vbae117-T2

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

24 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-29-06401fd5b220
Property and statement	Original source and location	Review and provenance
Diagram caption Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol. Individual claims	Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: version of record Retrieved: 2026-09-16T10:33:35.378Z	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `e95d4be70d32e61af5a92eda8ea66f25a2cc629e3f83e7b5241b13cde8bdb83b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram caption Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol. Individual claims	enbed-2024__vbae117_supplementary_data.pdf Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved 2026-09-16; sha256:a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32 Retrieved: 2026-09-16T21:05:56.086158+00:00	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32` Hash scope: Hash scope not separately documented; inspect source record Archive member: vbae117_supplementary_data.pdf Inspected artifact
Diagram steps Input: Genomic sequence. Evaluation: For the Nucleotide Transformer enhancer tasks, Supplementary Table 1 lists 14,968 training sequences and 400 test sequences of length 200. This set combines the original strong/weak/non-enhancer collection with 6,000 synthetic enhancers and 6,000 synthetic non-enhancers. Those counts do not describe the separate Genomic Benchmarks enhancer datasets. Readout: Classification accuracy is described for the benchmark comparison. Individual claims	Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: version of record Retrieved: 2026-09-16T10:33:35.378Z	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `e95d4be70d32e61af5a92eda8ea66f25a2cc629e3f83e7b5241b13cde8bdb83b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps Input: Genomic sequence. Evaluation: For the Nucleotide Transformer enhancer tasks, Supplementary Table 1 lists 14,968 training sequences and 400 test sequences of length 200. This set combines the original strong/weak/non-enhancer collection with 6,000 synthetic enhancers and 6,000 synthetic non-enhancers. Those counts do not describe the separate Genomic Benchmarks enhancer datasets. Readout: Classification accuracy is described for the benchmark comparison. Individual claims	enbed-2024__vbae117_supplementary_data.pdf Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved 2026-09-16; sha256:a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32 Retrieved: 2026-09-16T21:05:56.086158+00:00	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32` Hash scope: Hash scope not separately documented; inspect source record Archive member: vbae117_supplementary_data.pdf Inspected artifact
Diagram title Computational evaluation flow Individual claims	Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: version of record Retrieved: 2026-09-16T10:33:35.378Z	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `e95d4be70d32e61af5a92eda8ea66f25a2cc629e3f83e7b5241b13cde8bdb83b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Computational evaluation flow Individual claims	enbed-2024__vbae117_supplementary_data.pdf Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved 2026-09-16; sha256:a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32 Retrieved: 2026-09-16T21:05:56.086158+00:00	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32` Hash scope: Hash scope not separately documented; inspect source record Archive member: vbae117_supplementary_data.pdf Inspected artifact
Datasets Genomic Benchmarks includes human enhancer datasets; Nucleotide Transformer tasks include enhancer/non-enhancer and enhancer-strength labels. Individual claims	Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions Version: version of record Retrieved: 2026-09-16T10:33:35.378Z	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.facts.0.value` Source artifact SHA-256: `e95d4be70d32e61af5a92eda8ea66f25a2cc629e3f83e7b5241b13cde8bdb83b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Splits For the Nucleotide Transformer enhancer tasks, Supplementary Table 1 lists 14,968 training sequences and 400 test sequences of length 200. This set combines the original strong/weak/non-enhancer collection with 6,000 synthetic enhancers and 6,000 synthetic non-enhancers. Those counts do not describe the separate Genomic Benchmarks enhancer datasets. Individual claims	Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision Original source ↗ Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: version of record Retrieved: 2026-09-16T10:33:35.378Z	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.facts.1.value` Source artifact SHA-256: `e95d4be70d32e61af5a92eda8ea66f25a2cc629e3f83e7b5241b13cde8bdb83b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Splits For the Nucleotide Transformer enhancer tasks, Supplementary Table 1 lists 14,968 training sequences and 400 test sequences of length 200. This set combines the original strong/weak/non-enhancer collection with 6,000 synthetic enhancers and 6,000 synthetic non-enhancers. Those counts do not describe the separate Genomic Benchmarks enhancer datasets. Individual claims	enbed-2024__vbae117_supplementary_data.pdf Original source ↗ Supplement: Nucleotide Transformer benchmark dataset description and Table 1 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved 2026-09-16; sha256:a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32 Retrieved: 2026-09-16T21:05:56.086158+00:00	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.facts.1.value` Source artifact SHA-256: `a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32` Hash scope: Hash scope not separately documented; inspect source record Archive member: vbae117_supplementary_data.pdf Inspected artifact
Adaptation Task-level enhancer classification; exact task configuration must be selected from the evaluated benchmark collection. Individual claims	Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision Original source ↗ Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions Version: version of record Retrieved: 2026-09-16T10:33:35.378Z	source checked automated source review · 2026-09-16 Audit details Targeted full-paper and supplement review of the outstanding task fields, with original dataset metadata checked where accessible. Source-scoped omissions are explicit; no independent benchmark reproduction or numerical-result change. Field: `attributes.profile.facts.10.value` Source artifact SHA-256: `e95d4be70d32e61af5a92eda8ea66f25a2cc629e3f83e7b5241b13cde8bdb83b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

Release 2026-09-29-06401fd5b220 · Record review: needs review

3 source records and release history

Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Original source · version of record
Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision · Original source · version of record
enbed-2024__vbae117_supplementary_data.pdf · Original source · Retrieved 2026-09-16; sha256:a169bbebba2298e9c98c33c28053c1a0b42d0c3e88e5a8795934725c4408cc32

Download this release

Technical metadata and extraction receipts

Stable ID: reported-task-132da895d4c381

areas: dna-genomes
tasks: Enhancer classification
entity level: task
version: Not reported
task: Enhancer classification
scope note: Paper-specific evaluation task; protocol completeness requires further extraction.
benchmark research: review date: 2026-09-17; status: primary_comparison_tables_located; primary sources: evidence-expansion-enbed-2024-e95d4be7; inspected locators: Table 1.; XML table vbae117-T1; Table 2.; XML table vbae117-T2; searched queries: Understanding the natural language of DNA using encoder–decoder foundation models with byte-level precision 10.1093/bioadv/vbae117; gaps: complete numerical transcription and independent cell review: Full primary artifact and table inventory preserved; no new numeric row is published from this audit alone.; exact checkpoint hashes and per-method scored denominators: Table labels alone do not establish these fields; do not infer checkpoint or scored count from model name or dataset size.; claim scope: Dated primary-source discovery and protocol/table screening. Source checking does not mean experimental reproduction. Only separately extracted and independently reviewed numeric batches are publishable.
historical missing metadata: protocol version: not_reported_in_legacy_extract; split: not_reported_in_legacy_extract
metadata review scope: historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
legacy kinds: benchmark
entity classification: review date: 2026-09-17; rationale: This source-scoped record identifies the biological prediction task and holds its paper context. Preserve the existing task identity; exact split, model adaptation and scoring remain in linked evaluations or separate protocol records.; source ids: enbed-2024; source locator: Results §1.2.1; Methods §§2.5.1–2.5.2; cached text lines 18–19, 70–74; matching task comparison table/ablation captions; ambiguities: A paper- or suite-specific task may constrain some inputs or metrics; that alone does not make it interchangeable with a complete versioned protocol. No protocol equivalence is inferred.; Some legacy profile Entity type facts use the generic phrase computational evaluation protocol. That boilerplate is not sufficient to establish a single fixed protocol identity or to merge this task with another protocol record.

Related records

dataset: Genomic Benchmarks Mouse Enhancers
benchmark: ENBED: Enhancer classification
benchmark: ENBED (GRCh38): Enhancer classification

Suggest a correction