Configuration

mRNABERT

mRNABERT models complete mRNA sequences using dual tokenisation and protein-linked contrastive learning.

Sources

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Abstract (paragraph 1); Results/Overview of mRNABERT and benchmarks (paragraph 3)

1 evaluation · 1 metric row

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

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Abstract (paragraph 1); Methods/Model training (paragraph 3)

At a glance

Model type

Study-specific predictive method; this record is the paper-specific evaluated configuration.

Sources

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Abstract (paragraph 1); Methods/Model training (paragraph 3)

Biological inputs

mRNA sequences; protein correspondence supplies additional information during training

Sources

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Results/Applying mRNABERT to protein engineering tasks (paragraph 5); Results/Evaluating mRNABERT on CDS prediction tasks (paragraph 1)

Outputs

mRNA representations and task-specific design/property predictions

Sources

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Results/Predicting splice sites and alternative polyadenylation (paragraph 5); Discussion (paragraph 4)

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.

Evaluation setupMetricEvidence origin

Results grouped by the exact reported evaluation
Metric and finding	Coverage and uncertainty	Evidence
mRNABERT: translation-efficiency prediction Configuration: mRNABERTTask: translation-efficiency predictionDataset: human ultra-long mRNAs human translation-efficiency regression at 3066-nt input Author-reported evaluation · Evaluation metadata: needs review
0.669 R-squared Unit: unitless · Direction: unknown	Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported	source checkedmRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Table 2, mRNABERT (3066) row, Human R-squared column Source checking is not independent reproduction.

How it works

How the evaluated method works

The framework combines an mRNA language model with a cross-modality contrastive objective that incorporates information from corresponding protein sequences.

Sources

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Abstract (paragraph 1); Methods/Model training (paragraph 3)

What was evaluated

The linked evaluation record identifies mRNABERT: translation-efficiency prediction. Its dataset, split, adaptation and evidence origin remain attached to the reported results.

Sources

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-b2-mrnabert-2025

Strengths and limitations

Strengths and considerations

The evaluation includes UTR, CDS, RNA-binding-site and full-length mRNA-property tasks.
Sources
mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Abstract (paragraph 1); Results/Overview of mRNABERT and benchmarks (paragraph 4)

Limitations and conditions

The 3,066-nt evaluated configuration must not be conflated with an unconstrained full-transcript model; performance remains task- and adaptation-specific.
Sources
mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Results/Overview of mRNABERT and benchmarks (paragraph 4); Results/Applying mRNABERT to protein engineering tasks (paragraph 4)

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-13bd2a6c2d8178

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	Study-specific predictive method; this record is the paper-specific evaluated configuration. Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Abstract (paragraph 1); Methods/Model training (paragraph 3)
Architecture / procedure	The framework combines an mRNA language model with a cross-modality contrastive objective that incorporates information from corresponding protein sequences. Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Abstract (paragraph 1); Methods/Model training (paragraph 3)
Biological inputs	mRNA sequences; protein correspondence supplies additional information during training Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Results/Applying mRNABERT to protein engineering tasks (paragraph 5); Results/Evaluating mRNABERT on CDS prediction tasks (paragraph 1)
Outputs	mRNA representations and task-specific design/property predictions Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Results/Predicting splice sites and alternative polyadenylation (paragraph 5); Discussion (paragraph 4)
Parameters	Twelve transformer layers with hidden width 768 are specified. The inspected architecture section does not give a complete parameter total. · Not reported in inspected sources Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Methods/Model architecture (paragraph 1); Methods/Model evaluation (paragraph 2)
Known versions / configuration	3066-nt input Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Results/Evaluating the applicability using full-length mRNA sequences (paragraph 5); Table Tab2 (paragraph 1)
Training data / fitting	Approximately 36 million mature mRNAs were collected from NCBI nt, MG-RAST, GWH and MGnify, then filtered to approximately 18 million unique sequences for the curated corpus. Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Methods/Training datasets/Database construction and data collection (paragraph 2); Methods/Training datasets/Pre-processing the data (paragraph 2)
Context limits	This catalogue row is explicitly the 3,066-nt input configuration. Sources mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Results/Overview of mRNABERT and benchmarks (paragraph 2); Results/Capturing multi-dimensional biological information of mRNA (paragraph 3)
Access	Official study implementation and usage documentation: https://github.com/yyly6/mRNABERT/blob/893ccc920bb9be02a4677d14d96b03126da17689/README.md. This pinned documentation revision is not automatically the evaluated weight revision. Sources yyly6/mRNABERT README.md · README.md; installation, model download and usage instructions
Code licence	Apache 2.0 (study repository code at the cited revision; this does not establish every dependency or historical checkpoint licence). Sources yyly6/mRNABERT 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 sources Sources yyly6/mRNABERT 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.

19 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-17-d277315f7d76
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	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Abstract (paragraph 1); Methods/Model training (paragraph 3) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps ["mRNA sequences","mRNABERT","mRNA representations and task-specific design/property predictions"] Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Abstract (paragraph 1); Methods/Model training (paragraph 3) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Evaluated procedure (conceptual) Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Abstract (paragraph 1); Methods/Model training (paragraph 3) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Model type Study-specific predictive method; this record is the paper-specific evaluated configuration. Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Abstract (paragraph 1); Methods/Model training (paragraph 3) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Architecture / procedure The framework combines an mRNA language model with a cross-modality contrastive objective that incorporates information from corresponding protein sequences. Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Abstract (paragraph 1); Methods/Model training (paragraph 3) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` 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	yyly6/mRNABERT README.md Original source ↗ README.md; checkpoint/access documentation and licence scope Version: 893ccc920bb9be02a4677d14d96b03126da17689 Retrieved: 2026-09-16T19:54:19.395631+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: `bd50f1d7a71b1fd265b8b4c5590358d9bdabc4b6d897cd3b2ec6ba7ad5f7d2ff` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Biological inputs mRNA sequences; protein correspondence supplies additional information during training Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Results/Applying mRNABERT to protein engineering tasks (paragraph 5); Results/Evaluating mRNABERT on CDS prediction tasks (paragraph 1) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Outputs mRNA representations and task-specific design/property predictions Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Results/Predicting splice sites and alternative polyadenylation (paragraph 5); Discussion (paragraph 4) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Parameters Twelve transformer layers with hidden width 768 are specified. The inspected architecture section does not give a complete parameter total. Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Methods/Model architecture (paragraph 1); Methods/Model evaluation (paragraph 2) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Known versions / configuration 3066-nt input Individual claims	mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset Original source ↗ Results/Evaluating the applicability using full-length mRNA sequences (paragraph 5); Table Tab2 (paragraph 1) Version: journal full text in PMC Retrieved: 2026-09-16T10:38:57.558216+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.5.value` Source artifact SHA-256: `ff08ba895b7080446c08a930548b48a0041ae990c222ebb07e6ba7dcaf48ad44` 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

Release 2026-09-17-d277315f7d76 · Record review: needs review

3 source records and release history

mRNABERT: advancing mRNA sequence design with a universal language model and comprehensive dataset · Original source · journal full text in PMC
yyly6/mRNABERT README.md · Original source · 893ccc920bb9be02a4677d14d96b03126da17689
yyly6/mRNABERT LICENSE · Original source · 893ccc920bb9be02a4677d14d96b03126da17689

Download this release

Technical metadata and extraction receipts

Stable ID: reported-model-13bd2a6c2d8178

areas: rna-transcriptomes
entity level: method
version: 3066-nt input
reported name: mRNABERT
historical missing metadata: 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: mrnabert-2025; source locator: Abstract (paragraph 1); Methods/Model training (paragraph 3) | Abstract (paragraph 1); Results/Overview of mRNABERT and benchmarks (paragraph 3); 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.

Related records

model: mRNABERT: translation-efficiency prediction

Suggest a correction