Protocol

Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3)

Binary splice-altering classification on the SPiP benchmark variant set (99,601 variants), scored by sensitivity, specificity, F1 and balanced accuracy.

4 evaluations · 16 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.

4 recorded evaluations, 16 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

4 evaluations · 16 results. Different protocols are not a single leaderboard.

Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) · Dataset: SPiP benchmark variant set (99,601 variants)

Sorted by Balanced accuracy (maximum delta score > 0.5 (Legacy Ensemble N > 1)) (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	Balanced accuracy (maximum delta score > 0.5 (Legacy Ensemble N > 1))	F1 (maximum delta score > 0.5 (Legacy Ensemble N > 1))	Recall (maximum delta score > 0.5 (Legacy Ensemble N > 1))	Specificity (maximum delta score > 0.5 (Legacy Ensemble N > 1))
SpliceAI (Fortier et al. 2026)Evaluation	0.803 (best in column)	0.739 (best in column)	0.607 (best in column)	0.999 (best in column)
CI-SpliceAI (Fortier et al. 2026)Evaluation	0.798	0.725	0.596	0.999 (best in column)
OpenSpliceAI (Fortier et al. 2026)Evaluation	0.797	0.724	0.595	0.999 (best in column)
Legacy Ensemble (Fortier et al. 2026)Evaluation	0.752	0.289	0.547	0.956

All 16 result rows with coverage, uncertainty and sources

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

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

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

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.

Author-reported evaluations: 1
External evaluations: 3

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 Executed but unpublished runs and private review status are not included.

Null control

Proposed control: requires review

Training-set class prior where supervised fitting is permitted

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

Regularised classifier on simple permitted features, or protocol's conventional reference

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 2026-10-22-1b7cf027e4f6. 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.

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

Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Original source · PLOS One 21(5):e0348885, published 2026-05-13; PMC13170886 full-text XML

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: rna-splicing-20261011-fortier2026-protocol-spip

areas: dna-genomes
contexts: clinical_research
protocol: Binary classification using the authors' class_splice labels; maximum delta score above the standard 0.5 threshold is positive; Legacy Ensemble vote N > 1.
version: Table 3
source locator: Methods 'SPiP benchmark dataset'; Results 'SPiP benchmark results'; Table 3
limitations: Labels and variant sets are reused from earlier publications. CI-SpliceAI was trained on all human chromosomes, so no benchmark variant is held out from its training; ABCA4 lies on a test chromosome held out by both SpliceAI and OpenSpliceAI (Methods 'Technical overview of deep learning methods'; Discussion).; No uncertainty intervals are printed; significance is from two-proportion Z-tests on accuracy in the Results.

Related records

uses data: SPiP benchmark variant set (99,601 variants)
assessment: CI-SpliceAI on the SPiP benchmark variant set (99,601 variants)
assessment: Legacy Ensemble on the SPiP benchmark variant set (99,601 variants)
assessment: OpenSpliceAI on the SPiP benchmark variant set (99,601 variants)
assessment: SpliceAI on the SPiP benchmark variant set (99,601 variants)
assessed by: Select a patient-RNA splicing-variant validation workflow

Suggest a correction

Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: CI-SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.798 Balanced accuracy (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source CI-SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'CI-SpliceAI', column 'Balanced Accuracy'
Configuration: CI-SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.725 F1 (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source CI-SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'CI-SpliceAI', column 'F1'
Configuration: CI-SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.596 Recall (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source CI-SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'CI-SpliceAI', column 'Sensitivity'
Configuration: CI-SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.999 Specificity (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source CI-SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'CI-SpliceAI', column 'Specificity'
Configuration: Legacy Ensemble (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.752 Balanced accuracy (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Legacy Ensemble on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'Legacy Ensemble', column 'Balanced Accuracy'
Configuration: Legacy Ensemble (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.289 F1 (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Legacy Ensemble on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'Legacy Ensemble', column 'F1'
Configuration: Legacy Ensemble (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.547 Recall (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Legacy Ensemble on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'Legacy Ensemble', column 'Sensitivity'
Configuration: Legacy Ensemble (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.956 Specificity (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source Legacy Ensemble on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'Legacy Ensemble', column 'Specificity'
Configuration: OpenSpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.797 Balanced accuracy (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source OpenSpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'OpenSpliceAI', column 'Balanced Accuracy'
Configuration: OpenSpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.724 F1 (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source OpenSpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'OpenSpliceAI', column 'F1'
Configuration: OpenSpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.595 Recall (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source OpenSpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'OpenSpliceAI', column 'Sensitivity'
Configuration: OpenSpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.999 Specificity (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source OpenSpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'OpenSpliceAI', column 'Specificity'
Configuration: SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.803 Balanced accuracy (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'SpliceAI', column 'Balanced Accuracy'
Configuration: SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.739 F1 (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'SpliceAI', column 'F1'
Configuration: SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.607 Recall (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'SpliceAI', column 'Sensitivity'
Configuration: SpliceAI (Fortier et al. 2026)	Protocol: Splice-altering variant classification, SPiP benchmark variant set (99,601 variants) (Fortier et al. Table 3) Dataset: SPiP benchmark variant set (99,601 variants)	0.999 Specificity (maximum delta score > 0.5 (Legacy Ensemble N > 1)) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source SpliceAI on the SPiP benchmark variant set (99,601 variants) rna-splicing-20261011-fortier2026-protocol-spip Aggregation: Not reported Analyzing the performance of deep learning splice prediction algorithms (Fortier, Rudy and Scherer, PLOS One 2026) · Table 3, row 'SpliceAI', column 'Specificity'