rewire.it
Task

mRNABench GO: Gene Ontology term prediction

Gene Ontology term prediction. Scored with AUPRC (%) on mRNABench GO. A linear probe over frozen embeddings, scored as the mean over ten random seeds.

21 evaluations · 21 metric rows

Overview

Gene Ontology term prediction. Scored with AUPRC (%) on mRNABench GO. A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

Evaluation design

Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.

Benchmarks

These source-backed links do not make different protocols or scores interchangeable.

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Explore all linked results

Run instructions

No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.

Published comparisons

Explore the results reported under one evaluation protocol. Each figure keeps its source, dataset and metric together; it is not a ranking across studies. The pooled view gathers every source table that reports the same metric and names what it does not hold constant.

mRNABench GO: Gene Ontology term prediction

auprc (percent) · Higher values are better for this metric.

Every method mRNABench reports on Gene Ontology term prediction, scored with AUPRC (%) on mRNABench GO.

Evaluation protocol · mRNABench GO (mRNABench split)

  1. Naive Baseline · Method · Author-reported evaluation23.4
  2. Naive Mamba · Method · Author-reported evaluation26.6
  3. Supervised CNN · Method · Author-reported evaluation15.5
  4. 3UTRBERT · Configuration · Author-reported evaluation27.5
  5. AIDO.RNA · Configuration · Author-reported evaluation42.3
  6. DNABERT-S · Configuration · Author-reported evaluation31.8
  7. DNABERT2 · Configuration · Author-reported evaluation32.6
  8. ERNIE-RNA · Configuration · Author-reported evaluation34.1
  9. Evo1 · Configuration · Author-reported evaluation27.6
  10. Evo2 · Configuration · Author-reported evaluation46.7
  11. Helix-mRNA · Configuration · Author-reported evaluation25.7
  12. HyenaDNA · Configuration · Author-reported evaluation31.0
  13. NT · Configuration · Author-reported evaluation36.0
  14. Orthrus · Configuration · Author-reported evaluation42.9
  15. RNA-FM · Configuration · Author-reported evaluation32.2
  16. RNA-MSM · Configuration · Author-reported evaluation27.5
  17. RNABERT · Configuration · Author-reported evaluation23.8
  18. RNAErnie · Configuration · Author-reported evaluation27.8
  19. RiNALMo · Configuration · Author-reported evaluation34.7
  20. SpliceBERT · Configuration · Author-reported evaluation38.2
  21. UTR-LM · Configuration · Author-reported evaluation31.3

Source order is preserved. Plotted marks show point estimates; uncertainty, where reported, is retained in the printed values and table. Differences do not establish statistical significance.

mRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, column(GO)
Values, uncertainty and evidence
auprc: original source values
Tested entityPrinted valueUncertaintyEvidence
Naive Baseline · Method23.4 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Naive Baseline), column(GO)
Naive Mamba · Method26.6 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Naive Mamba), column(GO)
Supervised CNN · Method15.5 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Supervised CNN), column(GO)
3UTRBERT · Configuration27.5 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(3UTRBERT), column(GO)
AIDO.RNA · Configuration42.3 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(AIDO.RNA), column(GO)
DNABERT-S · Configuration31.8 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(DNABERT-S), column(GO)
DNABERT2 · Configuration32.6 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(DNABERT2), column(GO)
ERNIE-RNA · Configuration34.1 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(ERNIE-RNA), column(GO)
Evo1 · Configuration27.6 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Evo1), column(GO)
Evo2 · Configuration46.7 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Evo2), column(GO)
Helix-mRNA · Configuration25.7 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Helix-mRNA), column(GO)
HyenaDNA · Configuration31.0 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(HyenaDNA), column(GO)
NT · Configuration36.0 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(NT), column(GO)
Orthrus · Configuration42.9 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Orthrus), column(GO)
RNA-FM · Configuration32.2 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNA-FM), column(GO)
RNA-MSM · Configuration27.5 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNA-MSM), column(GO)
RNABERT · Configuration23.8 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNABERT), column(GO)
RNAErnie · Configuration27.8 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNAErnie), column(GO)
RiNALMo · Configuration34.7 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RiNALMo), column(GO)
SpliceBERT · Configuration38.2 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(SpliceBERT), column(GO)
UTR-LM · Configuration31.3 percentNot reportedAuthor-reported evaluation · source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(UTR-LM), column(GO)
Scope and limitations
  • Author-reported numbers, source checked but not independently reproduced.
  • Metrics alternate between AUPRC on a percentage scale and Pearson R, so figures in different columns are on different scales.
  • Each row is the best checkpoint of a model family, chosen by the authors, not the family's average.

Source transcription and grouping reviewed by automated source review on 2026-09-18. These experiments were not independently reproduced by rewire.

Tested entities and results

Release 2026-09-17-134cd1815de8 · 21 evaluations · 21 metric rows. Different protocols are not a single leaderboard. Where several source tables report the same metric, the published comparisons above offer a pooled view that names what it does not hold constant.

Results grouped by the exact reported evaluation
Metric and findingCoverage and uncertaintyEvidence
3UTRBERT on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

27.5% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(3UTRBERT), column(GO)

Source checking is not independent reproduction.

AIDO.RNA on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

42.3% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(AIDO.RNA), column(GO)

Source checking is not independent reproduction.

DNABERT-S on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

31.8% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(DNABERT-S), column(GO)

Source checking is not independent reproduction.

DNABERT2 on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

32.6% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(DNABERT2), column(GO)

Source checking is not independent reproduction.

ERNIE-RNA on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

34.1% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(ERNIE-RNA), column(GO)

Source checking is not independent reproduction.

Evo1 on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

27.6% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Evo1), column(GO)

Source checking is not independent reproduction.

Evo2 on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

46.7% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Evo2), column(GO)

Source checking is not independent reproduction.

Helix-mRNA on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

25.7% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Helix-mRNA), column(GO)

Source checking is not independent reproduction.

HyenaDNA on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

31.0% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(HyenaDNA), column(GO)

Source checking is not independent reproduction.

Naive Baseline on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

23.4% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Naive Baseline), column(GO)

Source checking is not independent reproduction.

Naive Mamba on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

26.6% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Naive Mamba), column(GO)

Source checking is not independent reproduction.

NT on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

36.0% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(NT), column(GO)

Source checking is not independent reproduction.

Orthrus on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

42.9% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Orthrus), column(GO)

Source checking is not independent reproduction.

RiNALMo on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

34.7% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RiNALMo), column(GO)

Source checking is not independent reproduction.

RNA-FM on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

32.2% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNA-FM), column(GO)

Source checking is not independent reproduction.

RNA-MSM on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

27.5% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNA-MSM), column(GO)

Source checking is not independent reproduction.

RNABERT on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

23.8% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNABERT), column(GO)

Source checking is not independent reproduction.

RNAErnie on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

27.8% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(RNAErnie), column(GO)

Source checking is not independent reproduction.

SpliceBERT on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

38.2% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(SpliceBERT), column(GO)

Source checking is not independent reproduction.

Supervised CNN on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

15.5% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(Supervised CNN), column(GO)

Source checking is not independent reproduction.

UTR-LM on mRNABench GO: Gene Ontology term prediction

A linear probe over frozen embeddings, scored as the mean over ten random seeds.

Author-reported evaluation · Evaluation metadata: source checked

31.3% auprc

Unit: percent · Direction: higher

Uncertainty: Not reported

Scored: Not reported · Eligible: Not reported

source checkedmRNABench: A curated benchmark for mature mRNA property and function prediction · Table 2, row(UTR-LM), column(GO)

Source checking is not 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.

1 evidence row matching the loaded filters

Claims, original sources and review scope · Release 2026-09-17-134cd1815de8
Property and statementOriginal source and locationReview and provenance
Relationship: part of

discovery-benchmark-mrnabench

Individual claims
mRNABench: A curated benchmark for mature mRNA property and function prediction

Original source ↗

Table 2, column(GO)

Version: preprint archived 2025-07-08
Retrieved: 2026-09-16T10:41:16.497221+00:00

source checked

automated source review · 2026-09-18

Audit details

Primary-source transcription with no human sign-off and no independent reproduction.

Field: links:part_of:discovery-benchmark-mrnabench

Claim: mrnabench-association-go

Source artifact SHA-256: 79f6264ee883535203c63a313547e7c57baa85585f76b42f8d899eb17fb7e600

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Sources and history

Release 2026-09-17-134cd1815de8 · Record review: source checked

1 source records and release historyDownload this release
Technical metadata and extraction receipts

Stable ID: mrnabench-task-go

areas
rna-transcriptomes
tasks
Gene Ontology term prediction
metric
AUPRC (%)
metric direction
higher
dataset
mRNABench GO
protocol
A linear probe over frozen embeddings, scored as the mean over ten random seeds.
source locator
Table 2, column(GO)
comparison panels
id: mrnabench-panel-go; title: mRNABench GO: Gene Ontology term prediction; protocol id: mrnabench-task-go; dataset id: mrnabench-dataset-mrnabench-go; metric: auprc; unit: percent; direction: higher; result ids: mrnabench-result-naive-baseline-go-auprc; mrnabench-result-naive-mamba-go-auprc; mrnabench-result-supervised-cnn-go-auprc; mrnabench-result-3utrbert-go-auprc; mrnabench-result-aido-rna-go-auprc; mrnabench-result-dnabert-s-go-auprc; mrnabench-result-dnabert2-go-auprc; mrnabench-result-ernie-rna-go-auprc; mrnabench-result-evo1-go-auprc; mrnabench-result-evo2-go-auprc; mrnabench-result-helix-mrna-go-auprc; mrnabench-result-hyenadna-go-auprc; mrnabench-result-nt-go-auprc; mrnabench-result-orthrus-go-auprc; mrnabench-result-rna-fm-go-auprc; mrnabench-result-rna-msm-go-auprc; mrnabench-result-rnabert-go-auprc; mrnabench-result-rnaernie-go-auprc; mrnabench-result-rinalmo-go-auprc; mrnabench-result-splicebert-go-auprc; mrnabench-result-utr-lm-go-auprc; source ids: expansion-p3-mrnabench-2025; source locator: Table 2, column(GO); context: Every method mRNABench reports on Gene Ontology term prediction, scored with AUPRC (%) on mRNABench GO.; caveats: Author-reported numbers, source checked but not independently reproduced.; Metrics alternate between AUPRC on a percentage scale and Pearson R, so figures in different columns are on different scales.; Each row is the best checkpoint of a model family, chosen by the authors, not the family's average.; review: method: automated_source_review; date: 2026-09-18
Related records

Suggest a correction