AUROC of ESM2 650M on human and viral proteins
99.7 AUROC
Methods
- Tested configuration
- ESM2 650M
- Protocol
- Held-out human-versus-virus protein classification (human-versus-viral protein classification)
- Dataset
- human and viral proteins
- Related family profiles
- ESM-2
- Procedure
- Compare protein origin classifiers; T 5 embeddings plus linear/tree models vs ESM 2 fine-tuning and simple controls. UniRef90 deduplication; proteins longer than1,600 residues excluded; UniRef50 clusters assigned80% training and20% test with no cluster shared. Separate four-fold error-analysis experiment not assigned toTable1.
- Evaluation
- ESM2 650M: human-versus-viral protein classification
- Coverage
- Not reported scored / Not reported eligible
- Uncertainty
- Not yet extracted
- Evidence
- Author-reported evaluation · source checkedProtein Language Models Expose Viral Immune Mimicry; Protein Language Models Expose Viral Immune Mimicry · Table 1, ESM2 650M row, AUC (%) column
A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: needs review. Source checked does not mean independently reproduced.
Reproduction
- Split
- UniRef90 deduplication; proteins longer than1,600 residues excluded; UniRef50 clusters assigned80% training and20% test with no cluster shared. Separate four-fold error-analysis experiment not assigned toTable1.
- Adaptation
- Paper-specific evaluated pipeline; exact checkpoint not inferred from label
- Scoring implementation
- Not reported
No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.
Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.
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 |
|---|---|---|
| Metric direction higher Individual claims | Protein Language Models Expose Viral Immune Mimicry Table 1, row ESM2 650M, column AUC (%); XML row7 column2 Version: version of record | source checked ["automated-source-review"] · 2026-09-17 author reported Audit detailsField: Claim: paper-claim-56208499359b816196 Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
| Printed value 99.67 Individual claims | Protein Language Models Expose Viral Immune Mimicry Table 1, ESM2 650M row, AUC (%) column Version: version of record | source checked ["independent-cell-check"] · 2026-09-16T10:33:57.274Z author reported Audit detailsExact row/header labels and numeric cell matched. Check verifies transcription, not experimental correctness. Field: Claim: claim-lit-b4-009 Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record Extraction artifact SHA-256: |
Sources and history
Release 2026-11-02-a7f03c9e55a5 · Record review: source checked
2 source records and release history
- Protein Language Models Expose Viral Immune Mimicry · Original source · version of record
- Protein Language Models Expose Viral Immune Mimicry · Original source · version of record
Technical metadata and extraction receipts
Stable ID: lit-b4-009
- areas
- proteins-complexes
- tasks
- human-versus-viral protein classification
- printed value
- 99.67
- numeric value
- 99.67
- metric
- auroc
- metric direction
- higher
- unit
- unitless
- source locator
- Table 1, ESM2 650M row, AUC (%) column
- review
- method: independent-cell-check; reviewer: rewire-deterministic-table-checker; reviewed at: 2026-09-16T10:33:57.274Z; evidence: Table 1, ESM2 650M row, AUC (%) column; cell: 99.67; artifact sha256: 15250af2f75f70e2b6a3725d00bc7e276ed9d2d4a54f0ae9d6eaabf6be13e4a1; retrieval url: https://www.ebi.ac.uk/europepmc/webservices/rest/PMC12474240/fullTextXML; method note: primary XML source; exact label and cell match; reviewer note: version v1; note: Exact row/header labels and numeric cell matched. Check verifies transcription, not experimental correctness.
- legacy id
- lit-b4-009
- missing metadata
- dataset version: reason: unextracted; model version: reason: unextracted; split: reason: unextracted; uncertainty: reason: unextracted
- unit detail
- printed x100