Model type
Study-specific predictive method; this record is the paper-specific evaluated configuration.
Pipeline
Stacking-Auto
Stacking-Auto is the sequence-based enhancer-location stage of the Hi-Enhancer framework.
Sources
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 1 Introduction (paragraph 6); 4 Discussion (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
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · Abstract (paragraph 1); 4 Discussion (paragraph 1)At a glance
Biological inputs
DNA sequences encoded with DNABERT-2
Sources
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model (paragraph 1); 4 Discussion/4.2 A novel strategy for enhancer localization (paragraph 1)Outputs
Enhancer predictions
Sources
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 1 Introduction (paragraph 3); 3 Results/3.3 Performance of Blending-KAN on cross-cell line prediction/3.3.4 Best performance of the five kinds of signal combinations (paragraph 2)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 |
|---|---|---|
| Stacking-Auto: enhancer prediction Two-stage Hi-Enhancer system; paper Table 2 method comparison Author-reported evaluation · Evaluation metadata: needs review | ||
| 80.50% accuracy Unit: percent · Direction: unknown | Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported | source checkedHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · Table 2, Ours (Stacking-Auto) row, Accuracy column Source checking is not independent reproduction. |
How it works
How the evaluated method works
DNABERT-2 extracts DNA sequence features and an AutoGluon stacking ensemble predicts enhancer labels. This is separate from the first-stage Blending-KAN predictor that uses epigenetic signals.
Sources
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · Abstract (paragraph 1); 4 Discussion (paragraph 1)What was evaluated
The linked evaluation record identifies Stacking-Auto: enhancer prediction. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-b4-003Strengths and limitations
Strengths and considerations
- Combines pretrained sequence features with multiple downstream learners.
Sources
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model (paragraph 2); 1 Introduction (paragraph 2)
Limitations and conditions
- The much higher epigenetic-signal-stage performance belongs to Blending-KAN and must not be presented as this sequence-based model’s result.
Sources
Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 3 Results/3.3 Performance of Blending-KAN on cross-cell line prediction/3.3.4 Best performance of the five kinds of signal combinations (paragraph 2); 3 Results/3.1 Performance of Blending-KAN on various signal combinations (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-0829aff5471d4bSpecifications
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.SourcesHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · Abstract (paragraph 1); 4 Discussion (paragraph 1) |
| Architecture / procedure | DNABERT-2 extracts DNA sequence features and an AutoGluon stacking ensemble predicts enhancer labels. This is separate from the first-stage Blending-KAN predictor that uses epigenetic signals.SourcesHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · Abstract (paragraph 1); 4 Discussion (paragraph 1) |
| Biological inputs | DNA sequences encoded with DNABERT-2SourcesHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model (paragraph 1); 4 Discussion/4.2 A novel strategy for enhancer localization (paragraph 1) |
| Outputs | Enhancer predictionsSourcesHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 1 Introduction (paragraph 3); 3 Results/3.3 Performance of Blending-KAN on cross-cell line prediction/3.3.4 Best performance of the five kinds of signal combinations (paragraph 2) |
| Parameters | An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sourcesSources (2)Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models; emanlee/Hi-Enhancer README.txt · 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.1 Preprocessing datasets; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model/2.1.2.1 Layer 1 of Blending-KAN; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model/2.1.2.2 Layer 2 of Blending-KAN; 2 Materials and methods/2.2 Localization of the boundaries of enhancers; 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model; 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.2 Locating enhancers; 3 Results/3.4 Comparison of Stacking-Auto with existing methods; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.txt at pinned repository revision |
| Known versions / configuration | Stacking-Auto is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sourcesSourcesHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | Stacking-Auto is fitted on the iEnhancer-2L benchmark dataset using DNABERT-2 embeddings and ten-fold cross-validated LightGBM meta-features. This differs from the HCT116/A549 data used for Blending-KAN.SourcesHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 3 Results/3.3 Performance of Blending-KAN on cross-cell line prediction (paragraph 1); 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model (paragraph 2) |
| Context limits | Enhancer localisation evaluates 77 sliding subsequences within each 4,000-bp region; the paper points to Supplementary Text S8 for exact window details.SourcesHi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · 4 Discussion (paragraph 1); 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.2 Locating enhancers (paragraph 1) |
| Access | Official study implementation and usage documentation: https://github.com/emanlee/Hi-Enhancer/blob/435bb1cc9ec2909d6ee551bb53c56f6a7bdde8fd/README.txt. This pinned documentation revision is not automatically the evaluated weight revision.Sourcesemanlee/Hi-Enhancer README.txt · README.txt; 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 sourcesSourcesemanlee/Hi-Enhancer README.txt · README.txt 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 sourcesSourcesemanlee/Hi-Enhancer README.txt · README.txt; 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.
20 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 | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models Abstract (paragraph 1); 4 Discussion (paragraph 1) Version: version of record | 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 ["DNA sequences encoded with DNABERT-2","Stacking-Auto","Enhancer predictions"] Individual claims | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models Abstract (paragraph 1); 4 Discussion (paragraph 1) Version: version of record | 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 | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models Abstract (paragraph 1); 4 Discussion (paragraph 1) Version: version of record | 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 | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models Abstract (paragraph 1); 4 Discussion (paragraph 1) Version: version of record | 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 DNABERT-2 extracts DNA sequence features and an AutoGluon stacking ensemble predicts enhancer labels. This is separate from the first-stage Blending-KAN predictor that uses epigenetic signals. Individual claims | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models Abstract (paragraph 1); 4 Discussion (paragraph 1) Version: version of record | 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 | emanlee/Hi-Enhancer README.txt README.txt; checkpoint/access documentation and licence scope Version: 435bb1cc9ec2909d6ee551bb53c56f6a7bdde8fd | 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 DNA sequences encoded with DNABERT-2 Individual claims | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model (paragraph 1); 4 Discussion/4.2 A novel strategy for enhancer localization (paragraph 1) Version: version of record | 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 Enhancer predictions Individual claims | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models 1 Introduction (paragraph 3); 3 Results/3.3 Performance of Blending-KAN on cross-cell line prediction/3.3.4 Best performance of the five kinds of signal combinations (paragraph 2) Version: version of record | 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 An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. Individual claims | emanlee/Hi-Enhancer README.txt 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.1 Preprocessing datasets; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model/2.1.2.1 Layer 1 of Blending-KAN; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model/2.1.2.2 Layer 2 of Blending-KAN; 2 Materials and methods/2.2 Localization of the boundaries of enhancers; 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model; 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.2 Locating enhancers; 3 Results/3.4 Comparison of Stacking-Auto with existing methods; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.txt at pinned repository revision Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 435bb1cc9ec2909d6ee551bb53c56f6a7bdde8fd | 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 |
| Parameters An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. Individual claims | Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.1 Preprocessing datasets; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model/2.1.2.1 Layer 1 of Blending-KAN; 2 Materials and methods/2.1 Blending-KAN for predicting enhancer regions/2.1.2 Blending-KAN model/2.1.2.2 Layer 2 of Blending-KAN; 2 Materials and methods/2.2 Localization of the boundaries of enhancers; 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.1 Stacking-Auto model; 2 Materials and methods/2.2 Localization of the boundaries of enhancers/2.2.2 Locating enhancers; 3 Results/3.4 Comparison of Stacking-Auto with existing methods; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.txt at pinned repository revision Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: version of record | 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
- Hi-Enhancer: a two-stage framework for prediction and localization of enhancers based on Blending-KAN and Stacking-Auto models · Original source · version of record
- emanlee/Hi-Enhancer README.txt · Original source · 435bb1cc9ec2909d6ee551bb53c56f6a7bdde8fd
Technical metadata and extraction receipts
Stable ID: reported-model-0829aff5471d4b
- areas
- dna-genomes
- entity level
- method
- version
- Not reported
- reported name
- Stacking-Auto
- 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: hi-enhancer-2025; source locator: Abstract (paragraph 1); 4 Discussion (paragraph 1) | 1 Introduction (paragraph 6); 4 Discussion (paragraph 1); ambiguities: This is the paper-specific pipeline identity; unspecified component checkpoints or implementation versions are not inferred from its name.