rewirebio.iobenchmarks
Protocol

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Rank score of each VEP from per-protein pairwise AUROC comparisons on 819 proteins (Table S8 rank_score).

85 evaluations · 85 results

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.

85 recorded evaluations, 85 metric rows. A comparison chart has not yet been validated for these results. The table retains the individual findings and their sources.

View coverage and remaining gaps across all benchmarks

Results

Results are available, but no reviewed comparison panel is linked in this release.

All evaluations

85 evaluations · 85 results. Different protocols are not a single leaderboard.

Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense · Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)

Sorted by Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each) (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 configurationPairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
SNPred (Livesey and Marsh 2025)Evaluation83.2% (best in column)
MetaRNN (Livesey and Marsh 2025)Evaluation82.2%
PHACTboost (Livesey and Marsh 2025)Evaluation80.7%
AlphaMissense (Livesey and Marsh 2025)Evaluation77.5%
VEST4 (Livesey and Marsh 2025)Evaluation77.1%
popEVE (Livesey and Marsh 2025)Evaluation76.6%
MutScore (Livesey and Marsh 2025)Evaluation76.6%
REVEL (Livesey and Marsh 2025)Evaluation76%
MOIpred_benign (Livesey and Marsh 2025)Evaluation74.3%
BayesDel_noAF (Livesey and Marsh 2025)Evaluation71.8%
CPT-1 (Livesey and Marsh 2025)Evaluation71.7%
GEMME (Livesey and Marsh 2025)Evaluation70.6%
iGEMME (Livesey and Marsh 2025)Evaluation70.4%
EVE (Livesey and Marsh 2025)Evaluation69.6%
MISTIC (Livesey and Marsh 2025)Evaluation69.4%
VARITY_R (Livesey and Marsh 2025)Evaluation69.3%
VESPA (Livesey and Marsh 2025)Evaluation68.7%
UNEECON (Livesey and Marsh 2025)Evaluation67.4%
gMVP (Livesey and Marsh 2025)Evaluation67%
ESCOTT (Livesey and Marsh 2025)Evaluation66.9%
BayesDel (Livesey and Marsh 2025)Evaluation66.9%
DeepSequence (Livesey and Marsh 2025)Evaluation65.9%
SaProt (Livesey and Marsh 2025)Evaluation65.5%
MutPred (Livesey and Marsh 2025)Evaluation65.1%
MutFormer (Livesey and Marsh 2025)Evaluation64%
CAPICE (Livesey and Marsh 2025)Evaluation63.7%
VARITY_ER (Livesey and Marsh 2025)Evaluation63.7%
ClinPred (Livesey and Marsh 2025)Evaluation62.4%
Rhapsody (Livesey and Marsh 2025)Evaluation61.7%
ESM-1v (Livesey and Marsh 2025)Evaluation61.4%
DEOGEN2 (Livesey and Marsh 2025)Evaluation60.8%
DeepSAV (Livesey and Marsh 2025)Evaluation59.1%
PHACT (Livesey and Marsh 2025)Evaluation58.3%
EVmutation_epistatic (Livesey and Marsh 2025)Evaluation58.1%
mvPPT (Livesey and Marsh 2025)Evaluation57.3%
Maverick_benign (Livesey and Marsh 2025)Evaluation57.1%
VESPAl (Livesey and Marsh 2025)Evaluation55.2%
SNAP2 (Livesey and Marsh 2025)Evaluation55%
mutationTCN (Livesey and Marsh 2025)Evaluation54.9%
Eigen (Livesey and Marsh 2025)Evaluation54.6%
ESM-1b (Livesey and Marsh 2025)Evaluation54.3%
SPRI (Livesey and Marsh 2025)Evaluation54.3%
InMeRF (Livesey and Marsh 2025)Evaluation53.4%
CADD_1.7 (Livesey and Marsh 2025)Evaluation52.1%
ESM2 (Livesey and Marsh 2025)Evaluation51.4%
MutationAssessor (Livesey and Marsh 2025)Evaluation51.4%
MVP (Livesey and Marsh 2025)Evaluation50.5%
SIGMA (Livesey and Marsh 2025)Evaluation50.4%
CONDEL (Livesey and Marsh 2025)Evaluation49.9%
MetaLR (Livesey and Marsh 2025)Evaluation49.1%
SIFT4G (Livesey and Marsh 2025)Evaluation49%
EVmutation_independent (Livesey and Marsh 2025)Evaluation48.9%
PROVEAN (Livesey and Marsh 2025)Evaluation48.6%
M-CAP (Livesey and Marsh 2025)Evaluation47.9%
DeMaSk (Livesey and Marsh 2025)Evaluation46.5%
SuSPect (Livesey and Marsh 2025)Evaluation46%
LASSIE (Livesey and Marsh 2025)Evaluation45.7%
MetaSVM (Livesey and Marsh 2025)Evaluation45.5%
MPC (Livesey and Marsh 2025)Evaluation45%
PonP2 (Livesey and Marsh 2025)Evaluation43.1%
Envision (Livesey and Marsh 2025)Evaluation42.4%
CADD (Livesey and Marsh 2025)Evaluation41.4%
Tranception (Livesey and Marsh 2025)Evaluation39.7%
PhD-SNPg (Livesey and Marsh 2025)Evaluation37.3%
LIST-S2 (Livesey and Marsh 2025)Evaluation35.3%
SIFT (Livesey and Marsh 2025)Evaluation34.8%
FathmmXF (Livesey and Marsh 2025)Evaluation32%
PrimateAI (Livesey and Marsh 2025)Evaluation31.6%
DANN (Livesey and Marsh 2025)Evaluation27.9%
AlphScore (Livesey and Marsh 2025)Evaluation25.3%
phyloP (Livesey and Marsh 2025)Evaluation23.4%
FATHMM (Livesey and Marsh 2025)Evaluation22.6%
COSMIS (Livesey and Marsh 2025)Evaluation22.2%
ProtGPT2 (Livesey and Marsh 2025)Evaluation21.2%
3Cnet (Livesey and Marsh 2025)Evaluation20.8%
fathmm-MKL (Livesey and Marsh 2025)Evaluation20.2%
LRT (Livesey and Marsh 2025)Evaluation18.8%
SiPhy (Livesey and Marsh 2025)Evaluation16.6%
GenoCanyon (Livesey and Marsh 2025)Evaluation14.3%
sequence_unet (Livesey and Marsh 2025)Evaluation14.1%
GERP++ (Livesey and Marsh 2025)Evaluation12.4%
BLOSUM62 (Livesey and Marsh 2025)Evaluation11.3%
Grantham (Livesey and Marsh 2025)Evaluation9.35%
phastCons (Livesey and Marsh 2025)Evaluation8.73%
fitCons (Livesey and Marsh 2025)Evaluation4.59%
All 85 result rows with coverage, uncertainty and sources
Filter evaluations

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configurationProtocol and datasetFindingEvidence and details
Configuration: 3Cnet (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
20.8% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 434/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ76)

Independent external evaluation · Source checked
Methods, coverage and source

3Cnet: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI76 (row '3Cnet', column 'rank_score')
Configuration: AlphaMissense (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
77.5% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 818/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ5)

Independent external evaluation · Source checked
Methods, coverage and source

AlphaMissense: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI5 (row 'AlphaMissense', column 'rank_score')
Configuration: AlphScore (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
25.3% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 796/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ71)

Independent external evaluation · Source checked
Methods, coverage and source

AlphScore: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI71 (row 'AlphScore', column 'rank_score')
Configuration: BayesDel_noAF (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
71.8% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 811/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ11)

Independent external evaluation · Source checked
Methods, coverage and source

BayesDel_noAF: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI11 (row 'BayesDel_noAF', column 'rank_score')
Configuration: BayesDel (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
66.9% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 814/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ22)

Independent external evaluation · Source checked
Methods, coverage and source

BayesDel: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI22 (row 'BayesDel', column 'rank_score')
Configuration: BLOSUM62 (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
11.3% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 819/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ83)

Independent external evaluation · Source checked
Methods, coverage and source

BLOSUM62: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI83 (row 'BLOSUM62', column 'rank_score')
Configuration: CADD_1.7 (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
52.1% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 812/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ45)

Independent external evaluation · Source checked
Methods, coverage and source

CADD_1.7: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI45 (row 'CADD_1.7', column 'rank_score')
Configuration: CADD (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
41.4% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 814/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ63)

Independent external evaluation · Source checked
Methods, coverage and source

CADD: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI63 (row 'CADD', column 'rank_score')
Configuration: CAPICE (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
63.7% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 813/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ27)

Independent external evaluation · Source checked
Methods, coverage and source

CAPICE: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI27 (row 'CAPICE', column 'rank_score')
Configuration: ClinPred (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
62.4% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 812/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ29)

Independent external evaluation · Source checked
Methods, coverage and source

ClinPred: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI29 (row 'ClinPred', column 'rank_score')
Configuration: CONDEL (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
49.9% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 809/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ50)

Independent external evaluation · Source checked
Methods, coverage and source

CONDEL: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI50 (row 'CONDEL', column 'rank_score')
Configuration: COSMIS (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
22.2% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 686/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ74)

Independent external evaluation · Source checked
Methods, coverage and source

COSMIS: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI74 (row 'COSMIS', column 'rank_score')
Configuration: CPT-1 (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
71.7% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 783/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ12)

Independent external evaluation · Source checked
Methods, coverage and source

CPT-1: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI12 (row 'CPT', column 'rank_score')
Configuration: DANN (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
27.9% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 814/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ70)

Independent external evaluation · Source checked
Methods, coverage and source

DANN: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI70 (row 'DANN', column 'rank_score')
Configuration: DeepSAV (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
59.1% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 813/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ33)

Independent external evaluation · Source checked
Methods, coverage and source

DeepSAV: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI33 (row 'DeepSAV', column 'rank_score')
Configuration: DeepSequence (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
65.9% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 200/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ23)

Independent external evaluation · Source checked
Methods, coverage and source

DeepSequence: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI23 (row 'DeepSequence', column 'rank_score')
Configuration: DeMaSk (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
46.5% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 814/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ56)

Independent external evaluation · Source checked
Methods, coverage and source

DeMaSk: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI56 (row 'DeMaSk', column 'rank_score')
Configuration: DEOGEN2 (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
60.8% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 795/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ32)

Independent external evaluation · Source checked
Methods, coverage and source

DEOGEN2: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI32 (row 'DEOGEN2', column 'rank_score')
Configuration: Eigen (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
54.6% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 738/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ41)

Independent external evaluation · Source checked
Methods, coverage and source

Eigen: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI41 (row 'Eigen', column 'rank_score')
Configuration: Envision (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
42.4% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 805/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ62)

Independent external evaluation · Source checked
Methods, coverage and source

Envision: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI62 (row 'Envision', column 'rank_score')
Configuration: ESCOTT (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
66.9% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 755/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ21)

Independent external evaluation · Source checked
Methods, coverage and source

ESCOTT: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI21 (row 'ESCOTT', column 'rank_score')
Configuration: ESM-1b (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
54.3% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 814/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ42)

Independent external evaluation · Source checked
Methods, coverage and source

ESM-1b: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI42 (row 'ESM-1b', column 'rank_score')
Configuration: ESM-1v (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
61.4% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 814/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ31)

Independent external evaluation · Source checked
Methods, coverage and source

ESM-1v: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI31 (row 'ESM-1v', column 'rank_score')
Configuration: ESM2 (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
51.4% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 819/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ46)

Independent external evaluation · Source checked
Methods, coverage and source

ESM2: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI46 (row 'ESM2', column 'rank_score')
Configuration: EVE (Livesey and Marsh 2025)Protocol: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense
Dataset: Livesey and Marsh 2025 clinical set: ClinVar pathogenic versus gnomAD v4.1 missense (819 proteins)
69.6% Pairwise win rate (average pairwise win rate against all other VEPs, per-protein AUROC on shared variants, ClinVar P/LP versus gnomAD v4.1 missense, proteins with at least 10 of each)
percent · higher

Uncertainty: Not reported by the source: Figure 3 error bars (standard error across comparisons) are not printed

Coverage: 738/819 proteins with at least 10 ClinVar P/LP and 10 gnomAD missense variants scored by this VEP scored ('proteins' column, cell CJ15)

Independent external evaluation · Source checked
Methods, coverage and source

EVE: Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Livesey and Marsh 2025 clinical benchmark: pairwise AUROC win rate, ClinVar pathogenic versus gnomAD missense

Aggregation: Not reported

Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8); Livesey and Marsh 2025, Genome Biology 26:104 (Europe PMC full text XML) · Supplementary Material 2, sheet 'Table S8', cell CI15 (row 'EVE', column 'rank_score')

Source checking is not independent reproduction. Release 2026-11-21-adc15377fab2.

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Explore all linked results

Baseline coverage

Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.

0 of 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.

No execution recipe linked to this protocol. Recipe availability does not establish a completed evaluation.

External evaluations
85

Literature evidence is not a Rewire measurement. Executed but unpublished runs and private review status are not included.

Null control

Proposed control: requires review

Select a task-valid null control after reviewing inputs and metric

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Conventional reference

Proposed control: requires review

Select an upstream conventional reference after reviewing the full protocol

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Protocol coverage CSV (gzip) · Model evaluation matrix (gzip) · Source table (gzip) · Release and checksums (gzip)

Coverage is derived from release 2026-11-21-adc15377fab2. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run instructions

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

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 2026-11-21-adc15377fab2
Property and statementOriginal source and locationReview and provenance

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

Sources and history

Release 2026-11-21-adc15377fab2 · Record review: source checked

2 source records and release historyDownload this release (gzip)
Technical metadata and record identifiers

Technical metadata for this record is tracked under stable ID vep-lit-20261014-protocol-livesey2025-clinical-auroc.

Source locator: Methods 'Spearman's correlation and rank score' and 'AUROC and AUPRC'; Supplementary Material 2 Legends (Table S8)

Related records

Suggest a correction