Model type
Protein sequence transformer; this record is the paper-specific evaluated configuration.
Configuration
ESM-2
This protein-language-model configuration is evaluated for viral RBD mutation properties under PRIME.
Sources
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Methods/Model architecture and scale (paragraph 1); Discussion (paragraph 1)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
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1)At a glance
Biological inputs
SARS-CoV-2 receptor-binding-domain sequences
Sources
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Results/Benchmark data collection for viral phenotype prediction (paragraph 3); Discussion (paragraph 7)Outputs
Binding-affinity and expression predictions
Sources
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Methods/Computational resources and reproducibility (paragraph 2); Results/Optimization of neural architectures for real-time deployment (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 |
|---|---|---|
| ESM-2: Mutated RBD binding prediction Frozen mean-pooled representation with downstream regression; position-stratified split. Independent external evaluation · Evaluation metadata: needs review | ||
| 0.0248 R² Unit: unitless · Direction: unknown | Uncertainty: ± 0.01 Scored: Not reported · Eligible: Not reported | source checkedPRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, ESM-2 8M / Mean / not fine-tuned row, Position-Stratified Split Binding R² column Source checking is not independent reproduction. |
How it works
How the evaluated method works
Protein embeddings feed downstream prediction heads, with frozen and fine-tuned encoder conditions explicitly separated.
Sources
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1)Underlying method and version boundaries
ESM-2 is a transformer protein language-model family. The official repository exposes residue embeddings, sequence-level pooling and models at several sizes; the study configuration determines which of these is evaluated.
Sources
facebookresearch/esm README.md · README.md; introduction, model description, pretrained-model and usage sections at pinned revisionWhat was evaluated
The linked evaluation record identifies ESM-2: Mutated RBD binding prediction. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-021Strengths and limitations
Strengths and considerations
- Contrasts random splitting with a position-stratified protocol that exposes leakage from closely related variants.
Sources
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Methods/Position-stratified validation protocol (paragraph 1); Methods/Position-stratified validation protocol/Leakage control (paragraph 1)
Limitations and conditions
- Random-split performance is not evidence of generalisation to unseen mutation sites; the row’s model size and adaptation must stay attached to its score.
Sources
PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Background (paragraph 5); Discussion (paragraph 2)
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-d0d5df2beb02b2Specifications
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 | Protein sequence transformer; this record is the paper-specific evaluated configuration.Sourcesfacebookresearch/esm README.md · README.md model description |
| Architecture / procedure | Protein embeddings feed downstream prediction heads, with frozen and fine-tuned encoder conditions explicitly separated.SourcesPRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1) |
| Biological inputs | SARS-CoV-2 receptor-binding-domain sequencesSourcesPRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Results/Benchmark data collection for viral phenotype prediction (paragraph 3); Discussion (paragraph 7) |
| Outputs | Binding-affinity and expression predictionsSourcesPRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Methods/Computational resources and reproducibility (paragraph 2); Results/Optimization of neural architectures for real-time deployment (paragraph 2) |
| Parameters | 8 million parameters, as identified for this rowSourcesPRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Abstract (paragraph 1); Methods/Clustering and phylogenetic analysis/Clustering (paragraph 1) |
| Known versions / configuration | 8MSourcesPRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table Tab1 (paragraph 1); Declarations/Consent for publication (paragraph 1) |
| Training data / fitting | 347,432 RBD sequences in the study; position-stratified partitions test unseen mutational sites.SourcesPRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Methods/Position-stratified validation protocol/Leakage control (paragraph 1); Background (paragraph 5) |
| Context limits | A maximum input/context length for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sourcesSources (2)PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space; facebookresearch/esm README.md · Results/Optimization of neural architectures for real-time deployment; Methods/Data collection and curation; Methods/Data collection and curation/Outbreak dataset; Methods/Data collection and curation/DMS dataset; Methods/Data collection and curation/BetaCov dataset; Methods/Model architecture and scale; Methods/Model architecture and scale/ESM-2 family; Methods/Model architecture and scale/ESM-C family; inspected for explicit maximum input length (dataset lengths and family-wide limits are not substituted); README.md at pinned repository revision |
| Access | Official upstream implementation and usage documentation: https://github.com/facebookresearch/esm/blob/2b369911bb5b4b0dda914521b9475cad1656b2ac/README.md. This pinned documentation revision is not automatically the evaluated weight revision.Sourcesfacebookresearch/esm 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).Sourcesfacebookresearch/esm 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 sourcesSourcesfacebookresearch/esm 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 | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1) 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 ["SARS-CoV-2 receptor-binding-domain sequences","ESM-2","Binding-affinity and expression predictions"] Individual claims | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1) 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 | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1) 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 Protein sequence transformer; this record is the paper-specific evaluated configuration. Individual claims | facebookresearch/esm README.md README.md model description Version: 2b369911bb5b4b0dda914521b9475cad1656b2ac | 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 Protein embeddings feed downstream prediction heads, with frozen and fine-tuned encoder conditions explicitly separated. Individual claims | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1) 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 | facebookresearch/esm README.md README.md; checkpoint/access documentation and licence scope Version: 2b369911bb5b4b0dda914521b9475cad1656b2ac | 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 SARS-CoV-2 receptor-binding-domain sequences Individual claims | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Results/Benchmark data collection for viral phenotype prediction (paragraph 3); Discussion (paragraph 7) 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 Binding-affinity and expression predictions Individual claims | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Methods/Computational resources and reproducibility (paragraph 2); Results/Optimization of neural architectures for real-time deployment (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 8 million parameters, as identified for this row Individual claims | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Abstract (paragraph 1); Methods/Clustering and phylogenetic analysis/Clustering (paragraph 1) 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 |
| Known versions / configuration 8M Individual claims | PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space Table Tab1 (paragraph 1); Declarations/Consent for publication (paragraph 1) 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 |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
3 source records and release history
- PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Original source · version of record
- facebookresearch/esm README.md · Original source · 2b369911bb5b4b0dda914521b9475cad1656b2ac
- facebookresearch/esm LICENSE · Original source · 2b369911bb5b4b0dda914521b9475cad1656b2ac
Technical metadata and extraction receipts
Stable ID: reported-model-d0d5df2beb02b2
- areas
- proteins-complexes
- entity level
- method
- version
- 8M
- reported name
- ESM-2
- 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: prime-2026; evidence-reported-base-esm-readme-md; source locator: Results/Optimization of neural architectures for real-time deployment (paragraph 1); Methods/Model architecture and scale (paragraph 1) | README.md model description | Methods/Model architecture and scale (paragraph 1); Discussion (paragraph 1); 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.