ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)

Overview

An explanatory profile has not yet been reviewed for this record.

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

Evaluations and results

8 evaluations · 8 results. Different protocols are not a single leaderboard.

Dataset: SARS-CoV-2 RBD mutant binding and expression measurements

Sorted by R² (ACE2 binding; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.

Tested configuration	Protocol and dataset	R² (ACE2 binding; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned)	R² (ACE2 binding; Random split over RBD mutants; the CLS token embedding, fine-tuned)	R² (RBD expression; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned)	R² (RBD expression; Random split over RBD mutants; the CLS token embedding, fine-tuned)	RMSE (ACE2 binding; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned)	RMSE (ACE2 binding; Random split over RBD mutants; the CLS token embedding, fine-tuned)	RMSE (RBD expression; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned)	RMSE (RBD expression; Random split over RBD mutants; the CLS token embedding, fine-tuned)
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME ACE2 binding regression, Position-stratified split holding out whole mutated positions, R²	-0.2294 ± 0.03	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME ACE2 binding regression, Position-stratified split holding out whole mutated positions, RMSE	Not reported	Not reported	Not reported	Not reported	1.9653 ± 0.03	Not reported	Not reported	Not reported
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME RBD expression regression, Position-stratified split holding out whole mutated positions, R²	Not reported	Not reported	-0.1928 ± 0.02	Not reported	Not reported	Not reported	Not reported	Not reported
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME RBD expression regression, Position-stratified split holding out whole mutated positions, RMSE	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	1.1756 ± 0.01	Not reported
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME ACE2 binding regression, Random split over RBD mutants, R²	Not reported	0.5897 ± 0.03	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME ACE2 binding regression, Random split over RBD mutants, RMSE	Not reported	Not reported	Not reported	Not reported	Not reported	1.2194 ± 0.04	Not reported	Not reported
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME RBD expression regression, Random split over RBD mutants, R²	Not reported	Not reported	Not reported	0.5302 ± 0.03	Not reported	Not reported	Not reported	Not reported
ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)Evaluation	PRIME RBD expression regression, Random split over RBD mutants, RMSE	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	0.6808 ± 0.02

All 8 result rows with coverage, uncertainty and sources

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME ACE2 binding regression, Position-stratified split holding out whole mutated positions, R² Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	-0.2294 ± 0.03 R² (ACE2 binding; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned) unitless · higher Uncertainty: SD 0.03 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: ACE2 binding R², Position-stratified split holding out whole mutated positions lit-protein-20261011-prime-protocol-position-stratified-binding-r2 Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Position-Stratified Split' / 'Binding' / 'R²'
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME ACE2 binding regression, Position-stratified split holding out whole mutated positions, RMSE Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	1.9653 ± 0.03 RMSE (ACE2 binding; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned) unit-unreported · lower Uncertainty: SD 0.03 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: ACE2 binding RMSE, Position-stratified split holding out whole mutated positions lit-protein-20261011-prime-protocol-position-stratified-binding-rmse Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Position-Stratified Split' / 'Binding' / 'RMSE'
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME RBD expression regression, Position-stratified split holding out whole mutated positions, R² Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	-0.1928 ± 0.02 R² (RBD expression; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned) unitless · higher Uncertainty: SD 0.02 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: RBD expression R², Position-stratified split holding out whole mutated positions lit-protein-20261011-prime-protocol-position-stratified-expression-r2 Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Position-Stratified Split' / 'Expression' / 'R²'
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME RBD expression regression, Position-stratified split holding out whole mutated positions, RMSE Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	1.1756 ± 0.01 RMSE (RBD expression; Position-stratified split holding out whole mutated positions; the CLS token embedding, fine-tuned) unit-unreported · lower Uncertainty: SD 0.01 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: RBD expression RMSE, Position-stratified split holding out whole mutated positions lit-protein-20261011-prime-protocol-position-stratified-expression-rmse Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Position-Stratified Split' / 'Expression' / 'RMSE'
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME ACE2 binding regression, Random split over RBD mutants, R² Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	0.5897 ± 0.03 R² (ACE2 binding; Random split over RBD mutants; the CLS token embedding, fine-tuned) unitless · higher Uncertainty: SD 0.03 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: ACE2 binding R², Random split over RBD mutants lit-protein-20261011-prime-protocol-random-binding-r2 Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Random Split' / 'Binding' / 'R²'
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME ACE2 binding regression, Random split over RBD mutants, RMSE Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	1.2194 ± 0.04 RMSE (ACE2 binding; Random split over RBD mutants; the CLS token embedding, fine-tuned) unit-unreported · lower Uncertainty: SD 0.04 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: ACE2 binding RMSE, Random split over RBD mutants lit-protein-20261011-prime-protocol-random-binding-rmse Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Random Split' / 'Binding' / 'RMSE'
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME RBD expression regression, Random split over RBD mutants, R² Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	0.5302 ± 0.03 R² (RBD expression; Random split over RBD mutants; the CLS token embedding, fine-tuned) unitless · higher Uncertainty: SD 0.03 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: RBD expression R², Random split over RBD mutants lit-protein-20261011-prime-protocol-random-expression-r2 Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Random Split' / 'Expression' / 'R²'
Configuration: ESM-2 8M, the CLS token embedding, fine-tuned (PRIME)	Protocol: PRIME RBD expression regression, Random split over RBD mutants, RMSE Dataset: SARS-CoV-2 RBD mutant binding and expression measurements	0.6808 ± 0.02 RMSE (RBD expression; Random split over RBD mutants; the CLS token embedding, fine-tuned) unit-unreported · lower Uncertainty: SD 0.02 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-2 8M CLS finetuned: RBD expression RMSE, Random split over RBD mutants lit-protein-20261011-prime-protocol-random-expression-rmse Aggregation: Not reported PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Table 1, row 'ESM-2 8M' / 'CLS' / '✓', columns 'Random Split' / 'Expression' / 'RMSE'

Source checking is not independent reproduction. Release 2026-10-22-1b7cf027e4f6.

Use this configuration

How it works, versions and access

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.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-22-1b7cf027e4f6
Property and statement	Original source and location	Review and provenance

No evidence rows match these filters. Choose another scope or clear the search.

How to read the evidence

Sources and history

Release 2026-10-22-1b7cf027e4f6 · Record review: source checked

1 source record and release history

PRIME: An evaluation framework for protein representation inference and generalization in viral mutation space · Original source · Version of record, BMC Genomics 2026 (PMC13425921 JATS full text)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: lit-protein-20261011-prime-config-esm-2-8m-cls-finetuned

areas: protein-function
contexts: research
method types: foundation_model
reported name: ESM-2 8M
version: ESM-2 8M
representation: the CLS token embedding
training setting: Fine-tuned on the RBD measurements
foundation model eligible: true
source locator: Table 1, row 'ESM-2 8M' / 'CLS' / '✓'

Related records