rewire.it
benchmark · suite

DART-Eval

DART-Eval examines whether DNA representations capture human gene regulation.

0 evaluations · 0 metric rows

At a glance

Explanatory profile: source reviewed · Automated source review, 2026-09-16. This does not change the review status of its results.

Data, procedure and scoring
PropertyDescription and evidence
Record typeTask suitekundajelab/DART-Eval official source · README: opening; Tasks 1–5 and their zero-shot, probing, fine-tuning and ab initio subsections
InputsHuman regulatory DNA and task-specific experimental annotationskundajelab/DART-Eval official source · README: opening; Tasks 1–5 and their zero-shot, probing, fine-tuning and ab initio subsections
Outputs and assessmentRegulatory task predictions in explicitly different adaptation regimeskundajelab/DART-Eval official source · README: opening; Tasks 1–5 and their zero-shot, probing, fine-tuning and ab initio subsections
DatasetsNot extracted or verified for this record.
OrganismsNot extracted or verified for this record.
AssaysNot extracted or verified for this record.
SplitsNot extracted or verified for this record.
AdaptationNot extracted or verified for this record.
BaselinesNot extracted or verified for this record.

How it works

Procedure overview

Conceptual overview of the cited procedure; consult the pinned source for executable settings.

Procedure overviewSelect regulatory task. Then: Choose adaptation regime. Then: Generate predictions. Then: Score task evidenceSelect regulatory taskChoose adaptation regimeGenerate predictionsScore task evidence
Read the diagram as text
  1. Select regulatory task
  2. Choose adaptation regime
  3. Generate predictions
  4. Score task evidence
kundajelab/DART-Eval official source · README: opening; Tasks 1–5 and their zero-shot, probing, fine-tuning and ab initio subsections

Procedure

Evaluate regulatory-element discrimination, motif footprinting, cell-type specificity, quantitative activity and variant effects using the supplied task resources. Distinguish zero-shot scoring, learned probes and fine-tuned models.

kundajelab/DART-Eval official source · README: opening; Tasks 1–5 and their zero-shot, probing, fine-tuning and ab initio subsections

Tested models and results

Release 2026-09-16-d74d282221a9 · 0 evaluations · 0 metric rows. Different protocols are not a single leaderboard.

No evaluations linked in this release.

Strengths and limitations

Strengths supported by sources

  • Includes increasing task difficulty and ab initio comparators.kundajelab/DART-Eval official source · README: opening; Tasks 1–5 and their zero-shot, probing, fine-tuning and ab initio subsections

Limitations and conditions

  • Results depend on the adaptation regime and biological task; zero-shot and fine-tuned scores cannot be pooled as the same evaluation.kundajelab/DART-Eval official source · README: opening; Tasks 1–5 and their zero-shot, probing, fine-tuning and ab initio subsections
Profile review details

Primary-source description checked by an automated research assistant. This is a profile review, not an independent execution or numerical reproduction.

Stable record: discovery-benchmark-dart-eval

Applicable tests and references

Applicability is distinct from a completed evaluation.

Sources and history

Release 2026-09-16-d74d282221a9 · Record review: discovered

Download this release
Technical metadata and extraction receipts

Stable ID: discovery-benchmark-dart-eval

areas
genomics
entity level
suite
missing metadata
dataset release: unextracted; metric implementation: unextracted; split manifest: unextracted; version: unextracted
scope note
Specialist molecular or omics evaluation; protocol details require review before numerical comparison.
task
Human regulatory DNA representation evaluation
version
Not reported
Related records

Suggest a correction