Configuration

GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)

Ablation: default model shown a 4 bp window at prediction time.

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

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

Sorted by AUPRC (maximum over three random seeds; zero-shot) (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	AUPRC (maximum over three random seeds; zero-shot)	AUPRC (mean over three random seeds; zero-shot)	AUROC (maximum over three random seeds; zero-shot)	AUROC (mean over three random seeds; zero-shot)	Odds ratio (rare (singletons) versus common (MAF > 5%) gnomAD variants in the tail of deleterious scores, threshold at 30 false discoveries, one-sided Fisher's exact test; all variant types, rare subsampled to the common count; maximum over three random seeds; zero-shot)	Odds ratio (rare (singletons) versus common (MAF > 5%) gnomAD variants in the tail of deleterious scores, threshold at 30 false discoveries, one-sided Fisher's exact test; all variant types, rare subsampled to the common count; mean over three random seeds; zero-shot)
GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)Evaluation	COSMIC frequent vs gnomAD common missense, AUPRC (GPN-MSA Supplementary Table 4), COSMIC frequent vs gnomAD common missense variants (GPN-MSA paper)	0.155 (best in column)	0.131 (best in column)	Not reported	Not reported	Not reported	Not reported
GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)Evaluation	OMIM pathogenic vs gnomAD common regulatory variants, AUPRC (GPN-MSA Supplementary Table 4), OMIM pathogenic vs gnomAD common regulatory variants (GPN-MSA paper)	0.053	0.052	Not reported	Not reported	Not reported	Not reported
GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)Evaluation	ClinVar Pathogenic vs gnomAD common missense, AUROC (GPN-MSA Supplementary Table 4), ClinVar Pathogenic vs gnomAD common missense variants (GPN-MSA paper)	Not reported	Not reported	0.943	0.942	Not reported	Not reported
GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)Evaluation	gnomAD rare vs common enrichment in the deleterious tail, odds ratio (GPN-MSA Supplementary Table 4), gnomAD rare (singletons) vs common variants, ablation set (GPN-MSA paper)	Not reported	Not reported	Not reported	Not reported	14.8	13.5

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: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: ClinVar Pathogenic vs gnomAD common missense, AUROC (GPN-MSA Supplementary Table 4) Dataset: ClinVar Pathogenic vs gnomAD common missense variants (GPN-MSA paper)	0.943 AUROC (maximum over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on ClinVar (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-clinvar-auroc Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'ClinVar' / 'max'
Configuration: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: ClinVar Pathogenic vs gnomAD common missense, AUROC (GPN-MSA Supplementary Table 4) Dataset: ClinVar Pathogenic vs gnomAD common missense variants (GPN-MSA paper)	0.942 AUROC (mean over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on ClinVar (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-clinvar-auroc Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'ClinVar' / 'mean'
Configuration: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: COSMIC frequent vs gnomAD common missense, AUPRC (GPN-MSA Supplementary Table 4) Dataset: COSMIC frequent vs gnomAD common missense variants (GPN-MSA paper)	0.155 AUPRC (maximum over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on COSMIC (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-cosmic-auprc Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'COSMIC' / 'max'
Configuration: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: COSMIC frequent vs gnomAD common missense, AUPRC (GPN-MSA Supplementary Table 4) Dataset: COSMIC frequent vs gnomAD common missense variants (GPN-MSA paper)	0.131 AUPRC (mean over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on COSMIC (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-cosmic-auprc Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'COSMIC' / 'mean'
Configuration: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: gnomAD rare vs common enrichment in the deleterious tail, odds ratio (GPN-MSA Supplementary Table 4) Dataset: gnomAD rare (singletons) vs common variants, ablation set (GPN-MSA paper)	14.8 Odds ratio (rare (singletons) versus common (MAF > 5%) gnomAD variants in the tail of deleterious scores, threshold at 30 false discoveries, one-sided Fisher's exact test; all variant types, rare subsampled to the common count; maximum over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on gnomAD (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-gnomad-odds-ratio Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'gnomAD' / 'max'
Configuration: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: gnomAD rare vs common enrichment in the deleterious tail, odds ratio (GPN-MSA Supplementary Table 4) Dataset: gnomAD rare (singletons) vs common variants, ablation set (GPN-MSA paper)	13.5 Odds ratio (rare (singletons) versus common (MAF > 5%) gnomAD variants in the tail of deleterious scores, threshold at 30 false discoveries, one-sided Fisher's exact test; all variant types, rare subsampled to the common count; mean over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on gnomAD (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-gnomad-odds-ratio Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'gnomAD' / 'mean'
Configuration: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: OMIM pathogenic vs gnomAD common regulatory variants, AUPRC (GPN-MSA Supplementary Table 4) Dataset: OMIM pathogenic vs gnomAD common regulatory variants (GPN-MSA paper)	0.053 AUPRC (maximum over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on OMIM (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-omim-auprc Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'OMIM' / 'max'
Configuration: GPN-MSA ablation 'Window size = 4' (Supplementary Table 4)	Protocol: OMIM pathogenic vs gnomAD common regulatory variants, AUPRC (GPN-MSA Supplementary Table 4) Dataset: OMIM pathogenic vs gnomAD common regulatory variants (GPN-MSA paper)	0.052 AUPRC (mean over three random seeds; zero-shot) unitless · higher Uncertainty: Not reported by the source: Only mean and max over seeds are printed Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Window size = 4 on OMIM (GPN-MSA Supplementary Table 4) gfm-guides-20261014-gpnmsa-protocol-ablation-omim-auprc Aggregation: Not reported GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Supplementary Table 4 (PDF page 23), row 'Window size = 4', column 'OMIM' / 'mean'

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

Use this configuration

How it works, versions and access

Strengths, limitations and unresolved questions

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

GPN-MSA: an alignment-based DNA language model for genome-wide variant effect prediction (bioRxiv v2) · Original source · bioRxiv v2, posted 2024-04-06

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: gfm-guides-20261014-gpnmsa-config-ablation-window-4

areas: dna-genomes
contexts: research
method types: foundation_model
reported name: Window size = 4
version: Ablation model trained by the authors for bioRxiv v2 Supplementary Table 4; not a released checkpoint
parameters: Default model, window 4 at variant effect prediction
access reuse: Ablation models are not listed among the published assets (gpn-msa.md); only songlab/gpn-msa-sapiens is released.
foundation model eligible: true
source locator: Supplementary Table 4 row 'Window size = 4'; Methods 'Ablation Study'

Related records

configuration of: GPN-MSA
system: Window size = 4 on ClinVar (GPN-MSA Supplementary Table 4)
system: Window size = 4 on COSMIC (GPN-MSA Supplementary Table 4)
system: Window size = 4 on gnomAD (GPN-MSA Supplementary Table 4)
system: Window size = 4 on OMIM (GPN-MSA Supplementary Table 4)

Suggest a correction