Model type
Study-specific predictive method; this record is the paper-specific evaluated configuration.
Pipeline
ARSENAL+ChromBPNet
This evaluated pipeline combines ARSENAL regulatory-DNA embeddings with ChromBPNet to predict chromatin accessibility.
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Results/ARSENAL embeddings improve supervised sequence models of chromatin accessibility (paragraph 2); Results/ARSENAL embeddings improve supervised sequence models of chromatin accessibility (paragraph 1)How it worksEvaluated procedure (conceptual)
Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings.
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2)At a glance
Biological inputs
Regulatory DNA sequence and task-specific supervised accessibility data
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Results/ARSENAL improves zero-shot motif discovery (paragraph 6); Methods/Supervised Variant Effect Prediction (paragraph 2)Outputs
Chromatin-accessibility predictions and derived regulatory-variant scores
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Methods/Zero-Shot Variant Effect Prediction (paragraph 1); Results/ARSENAL improves zero-shot variant effect prediction in regulatory QTL benchmarks (paragraph 3)limited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
Release 2026-09-17-d277315f7d76 · 1 evaluation · 1 metric row. Different protocols are not a single leaderboard.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| ARSENAL+ChromBPNet: regulatory-variant scoring Supervised ChromBPNet variant scoring with ARSENAL motif-discovery regularization Author-reported evaluation · Evaluation metadata: needs review | ||
| 0.896 AUROC Unit: fraction · Direction: unknown | Uncertainty: ±0.016 Scored: Not reported · Eligible: Not reported | source checkedShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Table 1, Yoruban LCL dsQTLs section, ARSENAL+ChromBPNet row, AUROC column Source checking is not independent reproduction. |
How it works
How the evaluated method works
ARSENAL is a short-context masked DNA language model with motif-discovery regularisation. Its embeddings provide additional inputs to a supervised ChromBPNet predictor; the linked result therefore evaluates the combined system.
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2)What was evaluated
The linked evaluation record identifies ARSENAL+ChromBPNet: regulatory-variant scoring. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-b4-001Strengths and limitations
Strengths and considerations
- Regulatory-enriched pretraining and motif regularisation target features that may be diluted in whole-genome pretraining.
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 1); Discussion (paragraph 3)
Limitations and conditions
- This source is a preprint, and improvements from supervised ChromBPNet integration must not be attributed to the pretrained encoder alone.
Sources
Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Methods/Supervised Variant Effect Prediction (paragraph 1); Methods/Supervised Model Training (paragraph 2)
Profile review details
Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction.
Stable record: reported-model-ee1ae8162c7d67Specifications
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.
| Property | Description and evidence |
|---|---|
| Model type | Study-specific predictive method; this record is the paper-specific evaluated configuration.SourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) |
| Architecture / procedure | ARSENAL is a short-context masked DNA language model with motif-discovery regularisation. Its embeddings provide additional inputs to a supervised ChromBPNet predictor; the linked result therefore evaluates the combined system.SourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) |
| Biological inputs | Regulatory DNA sequence and task-specific supervised accessibility dataSourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Results/ARSENAL improves zero-shot motif discovery (paragraph 6); Methods/Supervised Variant Effect Prediction (paragraph 2) |
| Outputs | Chromatin-accessibility predictions and derived regulatory-variant scoresSourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Methods/Zero-Shot Variant Effect Prediction (paragraph 1); Results/ARSENAL improves zero-shot variant effect prediction in regulatory QTL benchmarks (paragraph 3) |
| Parameters | Eight transformer blocks, hidden width 768, eight attention heads and feed-forward width 3,072; a combined ARSENAL–ChromBPNet parameter total is not reported in the inspected architecture section. · Not reported in inspected sourcesSourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Methods/Architecture and Pretraining (paragraph 3); Methods/Supervised Model Training (paragraph 2) |
| Known versions / configuration | ARSENAL+ChromBPNet is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sourcesSourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | ARSENAL pretraining uses approximately 2.3 million ENCODE candidate cis-regulatory elements; chromosomes 6 and 21 are validation and 5, 10, 14, 18, 20 and 22 are test.SourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 1); Methods/Pretraining Data (paragraph 1) |
| Context limits | ARSENAL encodes 350-bp cCRE windows; ChromBPNet integration supplies a separate supervised prediction procedure.SourcesShort-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Table T1 (paragraph 1); Methods/Sequence Generation (paragraph 7) |
| Access | Official study implementation and usage documentation: https://github.com/kundajelab/regulatory_lm/blob/2c4594f5b7649df3f1372238916f6d95bd31ba1d/README.md. This pinned documentation revision is not automatically the evaluated weight revision.Sourceskundajelab/regulatory_lm README.md · README.md; installation, model download and usage instructions |
| Code licence | No explicit code licence was established from the paper’s availability statement and inspected repository-root documentation. · Not reported in inspected sourcesSourceskundajelab/regulatory_lm README.md · README.md and repository-root licence-file search |
| Weights licence | The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. · Not reported in inspected sourcesSourceskundajelab/regulatory_lm README.md · README.md; checkpoint/access documentation and licence scope |
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.
19 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings. Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) Version: preprint version in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram steps ["Regulatory DNA sequence and task-specific supervised accessibility data","ARSENAL+ChromBPNet","Chromatin-accessibility predictions and derived regulatory-variant scores"] Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) Version: preprint version in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Evaluated procedure (conceptual) Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) Version: preprint version in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Model type Study-specific predictive method; this record is the paper-specific evaluated configuration. Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) Version: preprint version in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Architecture / procedure ARSENAL is a short-context masked DNA language model with motif-discovery regularisation. Its embeddings provide additional inputs to a supervised ChromBPNet predictor; the linked result therefore evaluates the combined system. Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) Version: preprint version in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Weights licence The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. Individual claims | kundajelab/regulatory_lm README.md README.md; checkpoint/access documentation and licence scope Version: 2c4594f5b7649df3f1372238916f6d95bd31ba1d | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Biological inputs Regulatory DNA sequence and task-specific supervised accessibility data Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Results/ARSENAL improves zero-shot motif discovery (paragraph 6); Methods/Supervised Variant Effect Prediction (paragraph 2) Version: preprint version in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Outputs Chromatin-accessibility predictions and derived regulatory-variant scores Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Methods/Zero-Shot Variant Effect Prediction (paragraph 1); Results/ARSENAL improves zero-shot variant effect prediction in regulatory QTL benchmarks (paragraph 3) Version: preprint version in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Parameters Eight transformer blocks, hidden width 768, eight attention heads and feed-forward width 3,072; a combined ARSENAL–ChromBPNet parameter total is not reported in the inspected architecture section. Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Methods/Architecture and Pretraining (paragraph 3); Methods/Supervised Model Training (paragraph 2) Version: preprint version in PMC | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Known versions / configuration ARSENAL+ChromBPNet is the comparison-table label; that label does not specify an immutable weight revision. Individual claims | Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. Version: preprint version in PMC | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
2 source records and release history
- Short-Context Regulatory DNA Language Models with Motif-Discovery Regularization · Original source · preprint version in PMC
- kundajelab/regulatory_lm README.md · Original source · 2c4594f5b7649df3f1372238916f6d95bd31ba1d
Technical metadata and extraction receipts
Stable ID: reported-model-ee1ae8162c7d67
- areas
- dna-genomes
- entity level
- method
- version
- Not reported
- reported name
- ARSENAL+ChromBPNet
- historical missing metadata
- version: not_reported_in_legacy_extract; checkpoint revision: not_reported_in_legacy_extract; training data: not_reported_in_legacy_extract; licence: 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
- model
- entity classification
- review date: 2026-09-17; rationale: This record identifies a composed analysis workflow with separately identifiable upstream models, representations or tools and a downstream prediction/scoring procedure. Results belong to that complete composition rather than to an upstream model alone.; source ids: arsenal-regulatory-dna-2026; source locator: Abstract (paragraph 1); Results/The ARSENAL model: targeted pretraining with motif discovery regularization (paragraph 2) | Results/ARSENAL embeddings improve supervised sequence models of chromatin accessibility (paragraph 2); Results/ARSENAL embeddings improve supervised sequence models of chromatin accessibility (paragraph 1); ambiguities: This is the paper-specific pipeline identity; unspecified component checkpoints or implementation versions are not inferred from its name.