rna-fm
Named representation or feature baseline with a task-specific linear probe; the complete evaluated pipeline is distinct from its underlying model.
Overview
Named representation or feature baseline with a task-specific linear probe; the complete evaluated pipeline is distinct from its underlying model.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluations and results
8 evaluations · 8 metric rows. Different protocols are not a single leaderboard.
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Pipeline: rna-fm | Protocol: mRNABench variant probes eCLIP: eCLIP binding site prediction Dataset subset: mRNABench eCLIP (mRNABench split) | 0.350 ± 0.002 auprc fraction · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.002; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes eCLIP: eCLIP binding site prediction Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 5 (XML T2), data row 37, XML tr 39, column 4 (eCLIP | AUPRC), model rna-fm |
| Pipeline: rna-fm | Protocol: mRNABench variant probes GO: Gene Ontology term prediction Dataset subset: mRNABench GO (mRNABench split) | 0.322 ± 0.004 auprc fraction · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.004; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes GO: Gene Ontology term prediction Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 6 (XML T3), data row 37, XML tr 39, column 5 (GO | AUPRC), model rna-fm |
| Pipeline: rna-fm | Protocol: mRNABench variant probes HL: mRNA half life Dataset subset: mRNABench HL (mRNABench split) | 0.472 ± 0.007 pearson_r correlation · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.007; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes HL: mRNA half life Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 6 (XML T3), data row 37, XML tr 39, column 2 (HL | Pearson R), model rna-fm |
| Pipeline: rna-fm | Protocol: mRNABench variant probes MRL-HL-Pair: Paired mean ribosome load and half life Dataset subset: mRNABench MRL-HL-Pair (mRNABench split) | 0.494 ± 0.071 pearson_r correlation · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.071; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes MRL-HL-Pair: Paired mean ribosome load and half life Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 6 (XML T3), data row 37, XML tr 39, column 3 (MRL-HL-Pair | Pearson R), model rna-fm |
| Pipeline: rna-fm | Protocol: mRNABench variant probes MRL-MPRA: Mean ribosome load on an MPRA library Dataset subset: mRNABench MRL MPRA (mRNABench split) | 0.487 ± 0.001 pearson_r correlation · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.001; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes MRL-MPRA: Mean ribosome load on an MPRA library Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 6 (XML T3), data row 37, XML tr 39, column 4 (MRL-MPRA | Pearson R), model rna-fm |
| Pipeline: rna-fm | Protocol: mRNABench variant probes MRL: Mean ribosome load Dataset subset: mRNABench MRL (mRNABench split) | 0.292 ± 0.017 pearson_r correlation · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.017; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes MRL: Mean ribosome load Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 5 (XML T2), data row 37, XML tr 39, column 5 (MRL | Pearson R), model rna-fm |
| Pipeline: rna-fm | Protocol: mRNABench variant probes Prot-Loc: Protein localisation Dataset subset: mRNABench Prot Loc (mRNABench split) | 0.322 ± 0.005 auprc fraction · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.005; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes Prot-Loc: Protein localisation Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 6 (XML T3), data row 37, XML tr 39, column 6 (Prot-Loc | AUPRC), model rna-fm |
| Pipeline: rna-fm | Protocol: mRNABench variant probes VEP: Variant effect prediction Dataset subset: mRNABench VEP (mRNABench split) | 0.257 ± 0.036 auprc fraction · higher Uncertainty: type: confidence_interval; confidence level: 0.95; reported half width: 0.036; aggregation: mean over ten random splits (table caption wording) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · source checkedMethods, coverage and sourcerna-fm on mRNABench variant probes VEP: Variant effect prediction Frozen transcript representations with task-specific linear probes, or named supervised/naive controls. Homology splits where possible; random splits for MRL-MPRA, MRL-HL-Pair and VEP. Reported mean over ten splits; Appendix C enumerates nine seeds. Table captions say ten random splits, while Methods specify homology splits where possible. This source ambiguity and the nine explicitly listed seeds are retained; exact split manifests remain unextracted. Appendix C uses micro-averaged multilabel metrics. Aggregation: Not reported mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 5 (XML T2), data row 37, XML tr 39, column 6 (VEP | AUPRC), model rna-fm |
Source checking is not independent reproduction. Release 2026-09-23-2b89723c6dd9.
Use this model
Strengths, limitations and unresolved questions
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.
2 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: uses model catalog-model-rna-fm Individual claims | mRNABench: A curated benchmark for mature mRNA property and function prediction Tables 5/6, row rna-fm; Appendix A (APP1), C (APP3), D (APP4) Version: preprint archived 2025-07-08 | source checked automated source review · 2026-09-23 Audit detailsSource-backed evaluated identity only; no independent reproduction. Field: Claim: mrnabench-variants-2025-method-rna-fm-catalog-model-rna-fm-identity-claim Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
| Relationship: uses model discovery-model-rna-fm Individual claims | mRNABench: A curated benchmark for mature mRNA property and function prediction Tables 5/6, row rna-fm; Appendix A (APP1), C (APP3), D (APP4) Version: preprint archived 2025-07-08 | source checked automated source review · 2026-09-23 Audit detailsSource-backed evaluated identity only; no independent reproduction. Field: Claim: mrnabench-variants-2025-method-rna-fm-discovery-model-rna-fm-identity-claim Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
Sources and history
View linked audit checks and correction history
Release 2026-09-23-2b89723c6dd9 · Record review: source checked
1 source records and release history
- mRNABench: A curated benchmark for mature mRNA property and function prediction · Original source · preprint archived 2025-07-08
Technical metadata and extraction receipts
Stable ID: mrnabench-variants-2025-method-rna-fm
- areas
- rna-transcriptomics
- source locator
- Tables 5/6, row rna-fm; Appendix A (APP1), C (APP3), D (APP4)
- missing metadata
- checkpoint revision: unreported; parameters: unextracted
Related records
- uses model: RNA-FM
- uses model: RNA-FM
- model: rna-fm on mRNABench variant probes eCLIP: eCLIP binding site prediction
- model: rna-fm on mRNABench variant probes GO: Gene Ontology term prediction
- model: rna-fm on mRNABench variant probes HL: mRNA half life
- model: rna-fm on mRNABench variant probes MRL: Mean ribosome load
- model: rna-fm on mRNABench variant probes MRL-HL-Pair: Paired mean ribosome load and half life
- model: rna-fm on mRNABench variant probes MRL-MPRA: Mean ribosome load on an MPRA library
- model: rna-fm on mRNABench variant probes Prot-Loc: Protein localisation
- model: rna-fm on mRNABench variant probes VEP: Variant effect prediction
- subject: rna-fm: uses model catalog-model-rna-fm
- subject: rna-fm: uses model discovery-model-rna-fm