Model type
Convolutional RNA folding model; this record is the paper-specific evaluated configuration.
Configuration
UFold
UFold is a learned RNA-secondary-structure comparator in the TU-Fold study.
Sources
RNA secondary structure prediction by conducting multi-class classifications · Materials and methods/Training method (paragraph 3); Results/TU-fold outperforms existing methods without using post-processing steps (paragraph 2)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
RNA secondary structure prediction by conducting multi-class classifications · Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3)At a glance
Biological inputs
RNA nucleotide sequences
Sources
RNA secondary structure prediction by conducting multi-class classifications · CRediT authorship contribution statement (paragraph 1); Discussion (paragraph 3)Outputs
RNA secondary-structure base pairs
Sources
RNA secondary structure prediction by conducting multi-class classifications · Materials and methods/Training method (paragraph 3); Materials and methods/Data augmentation (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 |
|---|---|---|
| UFold: RNA secondary structure Three-fold training and evaluation; source reports mean and standard deviation. Independent external evaluation · Evaluation metadata: needs review | ||
| 0.938 F1 Unit: unitless · Direction: unknown | Uncertainty: ± 0.004 standard deviation Scored: Not reported · Eligible: Not reported | source checkedRNA secondary structure prediction by conducting multi-class classifications · Table 2, UFold row, Overall F1 column Source checking is not independent reproduction. |
How it works
How the evaluated method works
RNA sequences become an image-like pairwise representation processed by a fully convolutional network, with base-pairing rules supporting structure prediction.
Sources
RNA secondary structure prediction by conducting multi-class classifications · Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3)Underlying method and version boundaries
UFold converts RNA sequences into an image-like representation and uses a fully convolutional network, trained on annotated structures and pairing rules, to predict secondary structure.
Sources
uci-cbcl/UFold README.md · README.md; introduction, model description, pretrained-model and usage sections at pinned revisionWhat was evaluated
The linked evaluation record identifies UFold: RNA secondary structure. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
RNA secondary structure prediction by conducting multi-class classifications · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-014Strengths and limitations
Strengths and considerations
- Provides an established convolutional comparator for the new multiclass formulation.
Sources
RNA secondary structure prediction by conducting multi-class classifications · Results/TU-fold outperforms existing methods without using post-processing steps (paragraph 5); Results/TU-fold outperforms existing methods without using post-processing steps (paragraph 7)
Limitations and conditions
- The comparator has its own training and post-processing conventions; TU-Fold augmentation must not be assigned to UFold.
Sources
RNA secondary structure prediction by conducting multi-class classifications · Introduction (paragraph 7); Introduction (paragraph 6)
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-e3abb0b9a2ec79Specifications
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 | Convolutional RNA folding model; this record is the paper-specific evaluated configuration.Sourcesuci-cbcl/UFold README.md · README.md model description |
| Architecture / procedure | RNA sequences become an image-like pairwise representation processed by a fully convolutional network, with base-pairing rules supporting structure prediction.SourcesRNA secondary structure prediction by conducting multi-class classifications · Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3) |
| Biological inputs | RNA nucleotide sequencesSourcesRNA secondary structure prediction by conducting multi-class classifications · CRediT authorship contribution statement (paragraph 1); Discussion (paragraph 3) |
| Outputs | RNA secondary-structure base pairsSourcesRNA secondary structure prediction by conducting multi-class classifications · Materials and methods/Training method (paragraph 3); Materials and methods/Data augmentation (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)RNA secondary structure prediction by conducting multi-class classifications; uci-cbcl/UFold README.md · Materials and methods/Training method; Materials and methods/Evaluation method; Materials and methods/Data augmentation; Materials and methods/Knowledge merge; Materials and methods/Experimental settings; Results/TU-fold outperforms existing methods without using post-processing steps; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.md at pinned repository revision |
| Known versions / configuration | UFold is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sourcesSourcesRNA secondary structure prediction by conducting multi-class classifications · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | The comparison evaluates UFold on the RNA8F collection under three random train/validation/test partitions. RNA8F combines RNAStrAlign, ArchiveII, bpRNA and CRW2, with CD-HIT filtering at 0.9. Exact UFold optimiser/checkpoint details are not stated in the comparison passage.SourcesRNA secondary structure prediction by conducting multi-class classifications · Materials and methods / Experimental settings; Table 2 |
| Context limits | The evaluated RNA8F collection contains sequences of at most 500 nucleotides. This is a dataset restriction, not a universal UFold inference limit.SourcesRNA secondary structure prediction by conducting multi-class classifications · Materials and methods / Experimental settings |
| Access | Official upstream implementation and usage documentation: https://github.com/uci-cbcl/UFold/blob/75bd9acc83826059682dfca9d3659df66b132cd1/README.md. This pinned documentation revision is not automatically the evaluated weight revision.Sourcesuci-cbcl/UFold README.md · README.md; installation, model download and usage instructions |
| Code licence | MIT (upstream repository code at the cited revision; this does not establish every dependency or historical checkpoint licence).Sourcesuci-cbcl/UFold LICENSE · LICENSE; complete licence text |
| 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 sourcesSourcesuci-cbcl/UFold 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.
21 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 | RNA secondary structure prediction by conducting multi-class classifications Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3) 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 ["RNA nucleotide sequences","UFold","RNA secondary-structure base pairs"] Individual claims | RNA secondary structure prediction by conducting multi-class classifications Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3) 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 | RNA secondary structure prediction by conducting multi-class classifications Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3) 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 Convolutional RNA folding model; this record is the paper-specific evaluated configuration. Individual claims | uci-cbcl/UFold README.md README.md model description Version: 75bd9acc83826059682dfca9d3659df66b132cd1 | 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 RNA sequences become an image-like pairwise representation processed by a fully convolutional network, with base-pairing rules supporting structure prediction. Individual claims | RNA secondary structure prediction by conducting multi-class classifications Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3) 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 | uci-cbcl/UFold README.md README.md; checkpoint/access documentation and licence scope Version: 75bd9acc83826059682dfca9d3659df66b132cd1 | 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 RNA nucleotide sequences Individual claims | RNA secondary structure prediction by conducting multi-class classifications CRediT authorship contribution statement (paragraph 1); Discussion (paragraph 3) 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 RNA secondary-structure base pairs Individual claims | RNA secondary structure prediction by conducting multi-class classifications Materials and methods/Training method (paragraph 3); Materials and methods/Data augmentation (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 | uci-cbcl/UFold README.md Materials and methods/Training method; Materials and methods/Evaluation method; Materials and methods/Data augmentation; Materials and methods/Knowledge merge; Materials and methods/Experimental settings; Results/TU-fold outperforms existing methods without using post-processing steps; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.md at pinned repository revision Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 75bd9acc83826059682dfca9d3659df66b132cd1 | 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 | RNA secondary structure prediction by conducting multi-class classifications Materials and methods/Training method; Materials and methods/Evaluation method; Materials and methods/Data augmentation; Materials and methods/Knowledge merge; Materials and methods/Experimental settings; Results/TU-fold outperforms existing methods without using post-processing steps; inspected for aggregate parameter count (component sizes are not added without an exact configuration); README.md 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
3 source records and release history
- RNA secondary structure prediction by conducting multi-class classifications · Original source · version of record
- uci-cbcl/UFold README.md · Original source · 75bd9acc83826059682dfca9d3659df66b132cd1
- uci-cbcl/UFold LICENSE · Original source · 75bd9acc83826059682dfca9d3659df66b132cd1
Technical metadata and extraction receipts
Stable ID: reported-model-e3abb0b9a2ec79
- areas
- rna-transcriptomes
- entity level
- method
- version
- Not reported
- reported name
- UFold
- 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 source-scoped entry preserves the method/configuration actually named in an evaluation. It is neither a global family identity nor proof of an immutable checkpoint; the linked evaluation retains adaptation, fitting and scoring details.; source ids: tu-fold-2025; evidence-reported-base-ufold-readme-md; source locator: Materials and methods/Training method (paragraph 1); Materials and methods/Training method (paragraph 3) | README.md model description | Materials and methods/Training method (paragraph 3); Results/TU-fold outperforms existing methods without using post-processing steps (paragraph 2); ambiguities: Configuration means the source-labelled evaluated identity. It does not establish missing checkpoint hashes, default settings or equivalence to same-named records in other papers.