Specificity of AlphaGenome Splice-Sites (Alavi 2026 v2) on OPA1 LOVD evaluation sets (49 splicing, 79 benign)
77 (97.5%) Specificity
Methods
- Tested configuration
- AlphaGenome Splice-Sites (Alavi 2026 v2)
- 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)
- Procedure
- Splice-defect classification of OPA1 variants with thresholds tuned on the same sets (Alavi Table III, evaluation sets)
- Evaluation
- AlphaGenome Splice-Sites on OPA1 LOVD evaluation sets (49 splicing, 79 benign) (Alavi 2026 v2)
- Coverage
- Not reported scored / Not reported eligible
- Uncertainty
- Not reported by the source
- Evidence
- Independent external evaluation · source checkedProbing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026, version 2) · Table III, row 'ALPHAGENOME SPLICE-SITES', column 'Specificity' (PDF page 6 of 12)
A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: source checked. Source checked does not mean independently reproduced.
Reproduction
- Split
- No split
- Adaptation
- zero-shot scores with thresholds grid-searched on the OPA1 evaluation sets
- Scoring implementation
- Equations 1 to 5 (type I error, type II error, accuracy, specificity, sensitivity); PPV
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.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Reported result 77 (97.5%) Individual claims | Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026, version 2) Table III, row 'ALPHAGENOME SPLICE-SITES', column 'Specificity' (PDF page 6 of 12) Version: bioRxiv preprint version 2, posted 2026-10-31; full-text PDF (12 pages) | source checked ["deterministic-table-parse","source-hash-verification","automated-source-review","ai-assisted-source-review"] · 2026-11-01 independent paper Audit detailsMatched by two parses of the pinned PDF: pdftotext -layout in build.py and pdftotext -raw in check_cells.py (cell-check.tsv, 56 of 56). Count, percentage, accuracy and PPV arithmetic checked for every row. Verdict in docs/reviews/2026-11-01-alavi-v2-rerun.md. Replaces version 1 result rna-splicing-20261011-alavi2026-result-opa1-evaluation-alphagenome-splice-scorer-specificity, which stays disputed and is not linked because the release drops records that link to disputed ones. Field: Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of the bioRxiv PDF, downloaded in a browser on 2026-11-01 (bioRxiv refused the scripted request). bioRxiv stamps the PDF when serving it (ModDate D:20261101144527 in this copy), so later retrievals differ. Extraction artifact SHA-256: |
Sources and history
Release 2026-11-02-a7f03c9e55a5 · Record review: source checked
1 source record and release history
- Probing Genomic Foundation Models with Splice-Variant Perturbations (Alavi, bioRxiv 2026, version 2) · Original source · bioRxiv preprint version 2, posted 2026-10-31; full-text PDF (12 pages)
Technical metadata and extraction receipts
Stable ID: rna-splicing-20261101-alavi2026-v2-result-opa1-evaluation-alphagenome-splice-sites-specificity
- metric
- specificity
- metric direction
- higher
- unit
- percent
- printed value
- 77 (97.5%)
- source locator
- Table III, row 'ALPHAGENOME SPLICE-SITES', column 'Specificity' (PDF page 6 of 12)
- review
- method: deterministic-table-parse; source-hash-verification; automated-source-review; ai-assisted-source-review; method note: pdftotext -layout (poppler 26.08.0) on the pinned PDF; value rows found by token pattern, 13 rows per table and row-label fragments asserted; printed cell text kept exactly; bracketed percentages checked against counts, and accuracy and PPV checked against the row's counts; the 126 non-AlphaGenome cells asserted equal to the stored version 1 results.; reviewer: claude; reviewer note: Claude agent for issue #213; no human review claimed. Extractor and reviewer were the same agent in the same session.; date: 2026-11-01; artifact sha256: 07bd32fffc1b31c222d43b0b18c213928e3fe111cc71c02fe2f95e64e6f7a798; retrieval url: https://www.biorxiv.org/content/10.64898/2026.10.02.756289v2.full.pdf; note: Matched by two parses of the pinned PDF: pdftotext -layout in build.py and pdftotext -raw in check_cells.py (cell-check.tsv, 56 of 56). Count, percentage, accuracy and PPV arithmetic checked for every row. Verdict in docs/reviews/2026-11-01-alavi-v2-rerun.md. Replaces version 1 result rna-splicing-20261011-alavi2026-result-opa1-evaluation-alphagenome-splice-scorer-specificity, which stays disputed and is not linked because the release drops records that link to disputed ones.
- numeric value
- 97.5
- missing metadata
- uncertainty: reason: unreported
- unit detail
- Cell prints the count of true negatives with the percentage in brackets; numeric_value is the percentage