Configuration

Borzoi (128-bp) (Alavi 2026)

BORZOI (128-BP) as configured in Alavi 2026.

Overview

Borzoi: Model trained to predict RNA-seq coverage from DNA sequence.

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

Evaluations and results

3 evaluations · 14 results. Different protocols are not a single leaderboard.

Sorted by Accuracy (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	Accuracy	False positive rate (type I error: false positives as a share of the benign set)	Precision	Proportion (type II error: false negatives as a share of the splicing set (false-negative rate))	Recall	Specificity	Proportion (share of 306 further OPA1 pathogenic substitutions called splice-altering)	Proportion (share of 69 OPA1 VUS substitutions called splice-altering)
Borzoi (128-bp) (Alavi 2026)Evaluation	Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets), OPA1 LOVD evaluation sets (49 splicing, 79 benign)	89.1% (best in column)	3 (3.8%) (best in column)	92.9% (best in column)	10 (20.4%) (best in column)	39 (79.6%) (best in column)	76 (96.2%) (best in column)	Not reported	Not reported
Borzoi (128-bp) (Alavi 2026)Evaluation	Spliceogenic versus benign classification on an independent 65-gene set with OPA1-derived thresholds (Alavi Table IV), Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)	80.9%	7 (6.5%)	87.5%	27 (35.5%)	49 (64.5%)	100 (93.5%)	Not reported	Not reported
Borzoi (128-bp) (Alavi 2026)Evaluation	Share of further OPA1 pathogenic substitutions and VUS called splice-altering (Alavi Table III, extended corpus), OPA1 extended LOVD corpus (306 pathogenic substitutions, 69 VUS)	Not reported	Not reported	Not reported	Not reported	Not reported	Not reported	70 (22.9%)	13 (18.8%)

All 14 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: Borzoi (128-bp) (Alavi 2026)	Protocol: Spliceogenic versus benign classification on an independent 65-gene set with OPA1-derived thresholds (Alavi Table IV) Dataset: Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)	80.9% Accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes) rna-splicing-20261011-alavi2026-protocol-multigene Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table IV, row 'BORZOI (128-BP)', column 'Accuracy' (PDF page 9 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Spliceogenic versus benign classification on an independent 65-gene set with OPA1-derived thresholds (Alavi Table IV) Dataset: Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)	87.5% Precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes) rna-splicing-20261011-alavi2026-protocol-multigene Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table IV, row 'BORZOI (128-BP)', column 'PPV' (PDF page 9 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Spliceogenic versus benign classification on an independent 65-gene set with OPA1-derived thresholds (Alavi Table IV) Dataset: Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)	49 (64.5%) Recall percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes) rna-splicing-20261011-alavi2026-protocol-multigene Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table IV, row 'BORZOI (128-BP)', column 'Sensitivity' (PDF page 9 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Spliceogenic versus benign classification on an independent 65-gene set with OPA1-derived thresholds (Alavi Table IV) Dataset: Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)	100 (93.5%) Specificity percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes) rna-splicing-20261011-alavi2026-protocol-multigene Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table IV, row 'BORZOI (128-BP)', column 'Specificity' (PDF page 9 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Spliceogenic versus benign classification on an independent 65-gene set with OPA1-derived thresholds (Alavi Table IV) Dataset: Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)	7 (6.5%) False positive rate (type I error: false positives as a share of the benign set) percent · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes) rna-splicing-20261011-alavi2026-protocol-multigene Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table IV, row 'BORZOI (128-BP)', column 'Type I error' (PDF page 9 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Spliceogenic versus benign classification on an independent 65-gene set with OPA1-derived thresholds (Alavi Table IV) Dataset: Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)	27 (35.5%) Proportion (type II error: false negatives as a share of the splicing set (false-negative rate)) percent · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes) rna-splicing-20261011-alavi2026-protocol-multigene Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table IV, row 'BORZOI (128-BP)', column 'Type II error' (PDF page 9 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets) Dataset: OPA1 LOVD evaluation sets (49 splicing, 79 benign)	89.1% Accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 LOVD evaluation sets (49 splicing, 79 benign) rna-splicing-20261011-alavi2026-protocol-opa1-evaluation Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'Accuracy' (PDF page 6 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets) Dataset: OPA1 LOVD evaluation sets (49 splicing, 79 benign)	92.9% Precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 LOVD evaluation sets (49 splicing, 79 benign) rna-splicing-20261011-alavi2026-protocol-opa1-evaluation Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'PPV' (PDF page 6 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets) Dataset: OPA1 LOVD evaluation sets (49 splicing, 79 benign)	39 (79.6%) Recall percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 LOVD evaluation sets (49 splicing, 79 benign) rna-splicing-20261011-alavi2026-protocol-opa1-evaluation Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'Sensitivity' (PDF page 6 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets) Dataset: OPA1 LOVD evaluation sets (49 splicing, 79 benign)	76 (96.2%) Specificity percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 LOVD evaluation sets (49 splicing, 79 benign) rna-splicing-20261011-alavi2026-protocol-opa1-evaluation Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'Specificity' (PDF page 6 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets) Dataset: OPA1 LOVD evaluation sets (49 splicing, 79 benign)	3 (3.8%) False positive rate (type I error: false positives as a share of the benign set) percent · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 LOVD evaluation sets (49 splicing, 79 benign) rna-splicing-20261011-alavi2026-protocol-opa1-evaluation Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'Type I error' (PDF page 6 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets) Dataset: OPA1 LOVD evaluation sets (49 splicing, 79 benign)	10 (20.4%) Proportion (type II error: false negatives as a share of the splicing set (false-negative rate)) percent · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 LOVD evaluation sets (49 splicing, 79 benign) rna-splicing-20261011-alavi2026-protocol-opa1-evaluation Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'Type II error' (PDF page 6 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Share of further OPA1 pathogenic substitutions and VUS called splice-altering (Alavi Table III, extended corpus) Dataset: OPA1 extended LOVD corpus (306 pathogenic substitutions, 69 VUS)	70 (22.9%) Proportion (share of 306 further OPA1 pathogenic substitutions called splice-altering) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 extended LOVD corpus (306 pathogenic substitutions, 69 VUS) rna-splicing-20261011-alavi2026-protocol-opa1-extended-positive-calls Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'Positive mutations' (PDF page 6 of 11)
Configuration: Borzoi (128-bp) (Alavi 2026)	Protocol: Share of further OPA1 pathogenic substitutions and VUS called splice-altering (Alavi Table III, extended corpus) Dataset: OPA1 extended LOVD corpus (306 pathogenic substitutions, 69 VUS)	13 (18.8%) Proportion (share of 69 OPA1 VUS substitutions called splice-altering) percent · unknown Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Borzoi (128-bp) on OPA1 extended LOVD corpus (306 pathogenic substitutions, 69 VUS) rna-splicing-20261011-alavi2026-protocol-opa1-extended-positive-calls Aggregation: Not reported Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Table III, row 'BORZOI (128-BP)', column 'Positive VUS' (PDF page 6 of 11)

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

Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026) · Original source · bioRxiv preprint version 1, posted 2026-10-09; full-text PDF (11 pages)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: rna-splicing-20261011-alavi2026-config-borzoi-128-bp

areas: dna-genomes
contexts: clinical_research
method types: supervised_machine_learning
reported name: BORZOI (128-BP)
foundation model eligible: false
version: Flashzoi implementation, johahi/borzoi-replicate-0 to -3
parameters: As 32-bp but at 128-bp resolution; LFC <= -0.10 or > +0.19
access reuse: Public weights (Hugging Face)
source locator: Methods C to H; Table III and IV row label

Related records

configuration of: Borzoi
system: Borzoi (128-bp) on Multi-gene spliceogenic test set (76 spliceogenic, 107 benign, 65 genes)
system: Borzoi (128-bp) on OPA1 LOVD evaluation sets (49 splicing, 79 benign)
system: Borzoi (128-bp) on OPA1 extended LOVD corpus (306 pathogenic substitutions, 69 VUS)

Suggest a correction