Configuration

SpliceAI2 tissue-specific fine-tune with a 147-RBP expression signature, 48 GTEx v8 tissues (Jaganathan et al. 2026)

SpliceAI2 tissue-specific fine-tune with a 147-RBP expression signature, 48 GTEx v8 tissues: Signature injection (bi-FiLM) before each of 20 MetaFormer blocks and a three-layer cross-attention decoder; 147 RBP expression values per tissue (TMM, log2 CPM, divided by 20); 1 kb flanks, input capped at 64 kb; 10 epochs.

1 evaluation · 3 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.

Evaluations and results

1 evaluation · 3 results. Different protocols are not a single leaderboard.

Protocol: Tissue-specific junction usage differences across GTEx tissue pairs, held-out chromosomes (SpliceAI2 paper, Pearson) · Dataset: GTEx v8 tissue-specific junction usage differences on held-out chromosomes (Jaganathan et al. 2026)

Sorted by Pearson (predicted versus observed logit junction usage difference, all tissue pairs, 14,712,948 comparisons) (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	Pearson (predicted versus observed logit junction usage difference, all tissue pairs, 14,712,948 comparisons)	Pearson (predicted versus observed logit junction usage difference, cerebellum versus blood, 13,770 junctions)	Pearson (predicted versus observed logit junction usage difference, cortex versus testis, 39,323 junctions)
SpliceAI2 tissue-specific fine-tune with a 147-RBP expression signature, 48 GTEx v8 tissues (Jaganathan et al. 2026)Evaluation	0.44	0.58	0.51

All 3 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: SpliceAI2 tissue-specific fine-tune with a 147-RBP expression signature, 48 GTEx v8 tissues (Jaganathan et al. 2026)	Protocol: Tissue-specific junction usage differences across GTEx tissue pairs, held-out chromosomes (SpliceAI2 paper, Pearson) Dataset: GTEx v8 tissue-specific junction usage differences on held-out chromosomes (Jaganathan et al. 2026)	0.44 Pearson (predicted versus observed logit junction usage difference, all tissue pairs, 14,712,948 comparisons) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source SpliceAI2 tissue-specific fine-tune on GTEx tissue-pair junction usage differences rna-splicing-20261013-spliceai2-protocol-gtex-tissue-specific Aggregation: Not reported A unified framework for quantitative splicing and transcript prediction (Jaganathan et al., bioRxiv 2026) · Main 'Modeling tissue-specific splicing' paragraph 2 '(Pearson r = 0.44; Fig. 2b)'; Fig. 2b panel 'All tissue pairs'
Configuration: SpliceAI2 tissue-specific fine-tune with a 147-RBP expression signature, 48 GTEx v8 tissues (Jaganathan et al. 2026)	Protocol: Tissue-specific junction usage differences across GTEx tissue pairs, held-out chromosomes (SpliceAI2 paper, Pearson) Dataset: GTEx v8 tissue-specific junction usage differences on held-out chromosomes (Jaganathan et al. 2026)	0.58 Pearson (predicted versus observed logit junction usage difference, cerebellum versus blood, 13,770 junctions) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source SpliceAI2 tissue-specific fine-tune on GTEx tissue-pair junction usage differences rna-splicing-20261013-spliceai2-protocol-gtex-tissue-specific Aggregation: Not reported A unified framework for quantitative splicing and transcript prediction (Jaganathan et al., bioRxiv 2026) · Main 'Modeling tissue-specific splicing' paragraph 2 '(13,770 splice junctions, Pearson r = 0.58)'; Fig. 2b panel 'Cerebellum vs blood'
Configuration: SpliceAI2 tissue-specific fine-tune with a 147-RBP expression signature, 48 GTEx v8 tissues (Jaganathan et al. 2026)	Protocol: Tissue-specific junction usage differences across GTEx tissue pairs, held-out chromosomes (SpliceAI2 paper, Pearson) Dataset: GTEx v8 tissue-specific junction usage differences on held-out chromosomes (Jaganathan et al. 2026)	0.51 Pearson (predicted versus observed logit junction usage difference, cortex versus testis, 39,323 junctions) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source SpliceAI2 tissue-specific fine-tune on GTEx tissue-pair junction usage differences rna-splicing-20261013-spliceai2-protocol-gtex-tissue-specific Aggregation: Not reported A unified framework for quantitative splicing and transcript prediction (Jaganathan et al., bioRxiv 2026) · Main 'Modeling tissue-specific splicing' paragraph 2 '(39,323 splice junctions, Pearson r = 0.51)'; Fig. 2b panel 'Cortex vs testis'

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

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-config-spliceai2-tissue-147rbp

areas: dna-genomes
contexts: research
method types: supervised_machine_learning
foundation model eligible: false
reported name: SpliceAI2 (tissue-specific)
version: Fine-tuned from the pretrained multispecies SpliceAI2; two replicates averaged
parameters: Signature injection (bi-FiLM) before each of 20 MetaFormer blocks and a three-layer cross-attention decoder; 147 RBP expression values per tissue (TMM, log2 CPM, divided by 20); 1 kb flanks, input capped at 64 kb; 10 epochs
training overlap: Fine-tuned on chromosomes 2, 4, 6, 8, 10 to 22 and X; chromosomes 1, 3, 5, 7 and 9 used for validation and evaluation. The pretrained encoder saw all human genes except chromosomes 21 and 22.
source locator: Methods 'Tissue-specific model architecture and fine-tuning'; Extended Data Fig. 3
missing metadata: checkpoint: reason: unreported; note: No released tissue-specific weights are named in the paper or README.
provenance note: SpliceAI2 tissue-specific fine-tune with a 147-RBP expression signature, 48 GTEx v8 tissues (Jaganathan et al. 2026), as evaluated in the SpliceAI2 paper.

Related records

configuration of: SpliceAI2
system: SpliceAI2 tissue-specific fine-tune on GTEx tissue-pair junction usage differences

Suggest a correction