Pipeline

DEBFold

DEBFold combines several RNA folding tools with a neural network and constrained thermodynamic optimisation.

Sources

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow (paragraph 1); Results and Discussion/DEBFold Is Robust against Different Thermodynamics-Constrained Optimization Algorithms (paragraph 1)

1 evaluation · 1 result

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

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1)

Overview

Model type

Convolutional neural network; this record is the paper-specific evaluated configuration.

Sources

Biological inputs

RNA sequences plus computed outputs of six folding procedures

Sources

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation (paragraph 5); Results and Discussion/DEBFold Is Robust against Different Thermodynamics-Constrained Optimization Algorithms (paragraph 1)

Outputs

RNA secondary structures

Sources

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Introduction (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1)

Access

The authors publish a web interface at https://cobis.bme.ncku.edu.tw/DEBFold/. The source establishes this access route; code and model-weight licence terms remain separate.

Sources

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Abstract and Introduction, DEBFold web interface

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

Evaluations and results

1 evaluation · 1 result. 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
Pipeline: DEBFold	Task: RNA secondary structure Dataset: DEBFold TestSetβ	55.7 Median F1 % · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source DEBFold: RNA secondary structure Median F1 on the prepared TestSetβ. Aggregation: Not reported DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Table 1, DEBFold row, TestSetβ F1 (%) column

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

Use this model

How it works, versions and access

How it works

How the evaluated method works

One-hot RNA and predictions from RNAfold, IPknot, MaxExpect, ProbKnot, RNAprob and Fold feed convolutional encoding/decoding and self-attention. The network estimates folding probabilities that become SHAPE-like constraints for free-energy minimisation.

Sources

What was evaluated

The linked evaluation record identifies DEBFold: RNA secondary structure. Its dataset, split, adaptation and evidence origin remain attached to the reported results.

Sources

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-015

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths and considerations

Integrates thermodynamic predictions with a learned constraint model and explicitly checks family-wise generalisation.
Sources
DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Results and Discussion/DEBFold Outperforms Previous Thermodynamics-Based RNA Structure Prediction Tools (paragraph 2); Abstract (paragraph 1)

Limitations and conditions

This is a multistage ensemble pipeline, with dependencies on six upstream tools; its result is not attributable to a single RNA language model.
Sources
DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Introduction (paragraph 5); Results and Discussion/Limitations of DEBFold (paragraph 1)

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-3f850c08d76410

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.

Inputs, outputs and configuration
Property	Description and evidence
Model type	Convolutional neural network; this record is the paper-specific evaluated configuration. Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1)
Architecture / procedure	One-hot RNA and predictions from RNAfold, IPknot, MaxExpect, ProbKnot, RNAprob and Fold feed convolutional encoding/decoding and self-attention. The network estimates folding probabilities that become SHAPE-like constraints for free-energy minimisation. Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1)
Biological inputs	RNA sequences plus computed outputs of six folding procedures Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation (paragraph 5); Results and Discussion/DEBFold Is Robust against Different Thermodynamics-Constrained Optimization Algorithms (paragraph 1)
Outputs	RNA secondary structures Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Introduction (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1)
Parameters	An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sources Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow; Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation; Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding; Methods and Data Sets/DEBFold Workflow/Model Training Hyperparameters; Methods and Data Sets/Family-Wise Processed RNA Structure Ground-Truth Data Set; Methods and Data Sets/Family-Wise Processed RNA Structure Ground-Truth Data Set/Contamination-Free Family-Wise Independent Test Set for Evaluating Existing Deep-Learning-Based Structure Prediction Tools; Methods and Data Sets/Family-Wise Processed RNA Structure Ground-Truth Data Set/PDB-Derived Source-Independent Test Set; Methods and Data Sets/Structure Prediction Evaluation Metrics; inspected for aggregate parameter count (component sizes are not added without an exact configuration)
Known versions / configuration	DEBFold is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sources Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label.
Training data / fitting	RNA-structure datasets with family-wise reserved test sets and a separate PDB-derived test set. Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Introduction (paragraph 5); Results and Discussion/DEBFold Has Better Generalization Performance than Existing Deep-Learning-Based Attempts (paragraph 1)
Context limits	Cross-validation inputs are padded to 512 nucleotides for batching; padding positions are excluded from loss and F1 calculations. Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation (paragraph 1); Methods and Data Sets/DEBFold Workflow/Model Training Hyperparameters (paragraph 1)
Access	The authors publish a web interface at https://cobis.bme.ncku.edu.tw/DEBFold/. The source establishes this access route; code and model-weight licence terms remain separate. Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Abstract and Introduction, DEBFold web interface
Code licence	No explicit code licence was established from the paper’s availability statement and inspected repository-root documentation. · Not reported in inspected sources Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/Family-Wise Processed RNA Structure Ground-Truth Data Set/Contamination-Free Family-Wise Independent Test Set for Evaluating Existing Deep-Learning-Based Structure Prediction Tools (paragraph 1); Conclusions (paragraph 1)
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 sources Sources DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation (paragraph 3); Methods and Data Sets/DEBFold Workflow/Model Training Hyperparameters (paragraph 1)

Evidence

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

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

Claims, original sources and review scope · Release 2026-09-29-06401fd5b220
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	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	source checked automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps RNA sequences plus computed outputs of six folding procedures DEBFold RNA secondary structures Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	source checked automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Evaluated procedure (conceptual) Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	source checked automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Model type Convolutional neural network; this record is the paper-specific evaluated configuration. Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	source checked automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.facts.0.value` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Architecture / procedure One-hot RNA and predictions from RNAfold, IPknot, MaxExpect, ProbKnot, RNAprob and Fold feed convolutional encoding/decoding and self-attention. The network estimates folding probabilities that become SHAPE-like constraints for free-energy minimisation. Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	source checked automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.facts.1.value` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
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	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation (paragraph 3); Methods and Data Sets/DEBFold Workflow/Model Training Hyperparameters (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	unreported automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.facts.10.value` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Biological inputs RNA sequences plus computed outputs of six folding procedures Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation (paragraph 5); Results and Discussion/DEBFold Is Robust against Different Thermodynamics-Constrained Optimization Algorithms (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	source checked automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.facts.2.value` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Outputs RNA secondary structures Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Introduction (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	source checked automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.facts.3.value` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Parameters An aggregate parameter total for this exact evaluated configuration is not established by the inspected sources. Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Methods and Data Sets/DEBFold Workflow; Methods and Data Sets/DEBFold Workflow/Stage 1: Structure Location Folding Probability Estimation; Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding; Methods and Data Sets/DEBFold Workflow/Model Training Hyperparameters; Methods and Data Sets/Family-Wise Processed RNA Structure Ground-Truth Data Set; Methods and Data Sets/Family-Wise Processed RNA Structure Ground-Truth Data Set/Contamination-Free Family-Wise Independent Test Set for Evaluating Existing Deep-Learning-Based Structure Prediction Tools; Methods and Data Sets/Family-Wise Processed RNA Structure Ground-Truth Data Set/PDB-Derived Source-Independent Test Set; Methods and Data Sets/Structure Prediction Evaluation Metrics; inspected for aggregate parameter count (component sizes are not added without an exact configuration) Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	unreported automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.facts.4.value` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Known versions / configuration DEBFold is the comparison-table label; that label does not specify an immutable weight revision. Individual claims	DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning Original source ↗ Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. Version: PMC11094721.1 Retrieved: 2026-09-16T10:41:16.509290+00:00	unreported automated source review · 2026-09-16 Audit 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. Field: `attributes.profile.facts.5.value` Source artifact SHA-256: `e8f960eafb7f00edfdd81d4fb75c6de838e9b872b7e18875fc7a5bff2a2f72b3` 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

1 source records and release history

DEBFold: Computational Identification of RNA Secondary Structures for Sequences across Structural Families Using Deep Learning · Original source · PMC11094721.1

Download this release

Technical metadata and extraction receipts

Stable ID: reported-model-3f850c08d76410

areas: rna-transcriptomes
entity level: method
version: Not reported
reported name: DEBFold
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: debfold-2024; source locator: Methods and Data Sets/DEBFold Workflow/Stage 2: Score-Constrained Optimization Folding (paragraph 1); Methods and Data Sets/DEBFold Workflow (paragraph 1) | Methods and Data Sets/DEBFold Workflow (paragraph 1); Results and Discussion/DEBFold Is Robust against Different Thermodynamics-Constrained Optimization Algorithms (paragraph 1); ambiguities: This is the paper-specific pipeline identity; unspecified component checkpoints or implementation versions are not inferred from its name.

Related records

model: DEBFold: RNA secondary structure

Suggest a correction