Result

AUROC (all variants; essential splice variants excluded) of SpliceAI 1.3.1, position-resolved delta at the effect position, masking off (Jaganathan et al. 2026) on GTEx v8 sQTL fine-mapped and null SNVs near junction ends (Jaganathan et al. 2026)

0.71 AUROC (all variants; essential splice variants excluded)

Methods

Tested configuration: SpliceAI 1.3.1, position-resolved delta at the effect position, masking off (Jaganathan et al. 2026)
Protocol: GTEx v8 fine-mapped sQTL SNVs against non-prioritised SNVs near junction ends (SpliceAI2 paper, auROC)
Dataset: GTEx v8 sQTL fine-mapped and null SNVs near junction ends (Jaganathan et al. 2026)
Procedure: GTEx v8 fine-mapped sQTL SNVs against non-prioritised SNVs near junction ends (SpliceAI2 paper, auROC)
Evaluation: SpliceAI on GTEx sQTL (auroc, SpliceAI2 paper)
Coverage: Not reported scored / Not reported eligible
Uncertainty: Not reported by the source
Evidence: Author-reported evaluation · source checkedA unified framework for quantitative splicing and transcript prediction (Jaganathan et al., bioRxiv 2026) · Extended Data Fig. 2 (page 52), panel 'GTEx sQTL (auROC)', row 'SpliceAI', column 'All'

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: None (pretrained scoring)
Adaptation: Not reported
Scoring implementation: auROC

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

Claims, original sources and review scope · Release 2026-10-22-1b7cf027e4f6
Property and statement	Original source and location	Review and provenance
Reported result 0.71 Individual claims	A unified framework for quantitative splicing and transcript prediction (Jaganathan et al., bioRxiv 2026) Original source ↗ Extended Data Fig. 2 (page 52), panel 'GTEx sQTL (auROC)', row 'SpliceAI', column 'All' Version: bioRxiv preprint version 1, posted 2026-10-11; full-text PDF (57 pages) Retrieved: 2026-10-13T12:40:34Z	source checked ["pdf-text-parse","independent-cell-check","rendered-source-visual-check"] · 2026-10-13 author reported Audit details Checked against the pinned source by the independent reviewer; verdict and correction list in docs/reviews/use-cases/spliceai2-2026-10-13.md and this batch's review.json. Developer-reported value, not reproduced. Extraction method: Deterministic parse of `pdftotext -layout` (poppler) text of page 52 of the pinned PDF: panel titles asserted in order, model row labels asserted, 17 values per row; column labels (rotated in the figure) read from a 130 dpi render of page 52. 'All' column cross-checked against the Fig. 1d bar labels and the main-text statements (extract/build_batch.py). Field: `attributes.printed_value` Source artifact SHA-256: `f8f3db2c3943d4827b670bc2e962e0cec8fbe96e0b1acf9d4a9b9f0c7a78c06b` Hash scope: SHA-256 of one retrieval of the bioRxiv PDF. bioRxiv stamps the PDF when serving it (XMP ModifyDate, MetadataDate, DocumentID, InstanceID, Info ModDate and the second trailer ID), so later retrievals differ. A reviewer retrieval on 2026-10-13 of the same length (2,679,664 bytes) had SHA-256 407f5884427dda0226e0ce1f022e4479d3ff2b243da18bcc691eb345b0e04a6a and ModDate D:20261013063115-07'00'; its bytes differ only in those fields and its pdftotext text layer is identical to the pinned copy. Inspected artifact Extraction artifact SHA-256: `f8f3db2c3943d4827b670bc2e962e0cec8fbe96e0b1acf9d4a9b9f0c7a78c06b` Extraction artifact

How to read the evidence

Sources and history

Release 2026-10-22-1b7cf027e4f6 · Record review: source checked

1 source record and release history

A unified framework for quantitative splicing and transcript prediction (Jaganathan et al., bioRxiv 2026) · Original source · bioRxiv preprint version 1, posted 2026-10-11; full-text PDF (57 pages)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: rna-splicing-20261013-spliceai2-result-gtex-sqtl-classification-spliceai-position-resolved-auroc-all

metric: auroc
metric direction: higher
unit: unitless
printed value: 0.71
numeric value: 0.71
source locator: Extended Data Fig. 2 (page 52), panel 'GTEx sQTL (auROC)', row 'SpliceAI', column 'All'
review: method: pdf-text-parse; independent-cell-check; rendered-source-visual-check; reviewer: claude; reviewer note: Claude (Opus 5.5) research agent, independent reviewer; did not extract this batch; no human review claimed; date: 2026-10-13; artifact sha256: f8f3db2c3943d4827b670bc2e962e0cec8fbe96e0b1acf9d4a9b9f0c7a78c06b; retrieval url: https://www.biorxiv.org/content/10.64898/2026.10.08.757673v1.full.pdf; method note: Reviewer's own pdftotext -layout parse of Extended Data Fig. 2 (page 52): the four model rows of the top block, 17 values each, matched one-to-one to the 68 stored results by model, panel and stratum; page rendered at 110 dpi and column labels read (All, Exon, Near intronic, Deep intronic; All, Exon, Intron). The All column agrees with the Fig. 1d bar labels on the rendered page 11.; note: Checked against the pinned source by the independent reviewer; verdict and correction list in docs/reviews/use-cases/spliceai2-2026-10-13.md and this batch's review.json. Developer-reported value, not reproduced. Extraction method: Deterministic parse of `pdftotext -layout` (poppler) text of page 52 of the pinned PDF: panel titles asserted in order, model row labels asserted, 17 values per row; column labels (rotated in the figure) read from a 130 dpi render of page 52. 'All' column cross-checked against the Fig. 1d bar labels and the main-text statements (extract/build_batch.py).
missing metadata: uncertainty: reason: unreported
metric qualifier: all variants; essential splice variants excluded

Related records

evaluation: SpliceAI on GTEx sQTL (auroc, SpliceAI2 paper)

Suggest a correction