SpliceAI predicts how sequence variants alter splice-site usage. Its variant annotation tool combines reference sequence with gene annotation.
Explanatory profile: source reviewed · Automated source review, 2026-09-16. This does not change the review status of its results.
| Property | Description and evidence |
|---|---|
| Inputs | VCF, reference FASTA and gene annotationillumina/SpliceAI official source · README.md: package description, License, Usage and Frequently Asked Questions |
| Outputs | Acceptor/donor gain and loss delta scoresillumina/SpliceAI official source · README.md: package description, License, Usage and Frequently Asked Questions |
| Model type | Not extracted or verified for this record. |
| Known versions | Not extracted or verified for this record. |
| Training data | Not extracted or verified for this record. |
| Context limits | Not extracted or verified for this record. |
| Access | Not extracted or verified for this record. |
| Code licence | Not extracted or verified for this record. |
| Weights licence | Not extracted or verified for this record. |
Schematic of the documented input, computation and output; not an executable configuration.
SpliceAI uses a residual convolutional network with dilated filters to integrate sequence context. It predicts donor, acceptor and non-splice-site probabilities along the sequence; comparing alleles converts those predictions into variant scores. The output is not tissue-specific.
illumina/SpliceAI official source; tkzeng/Pangolin official source · SpliceAI README and linked Jaganathan et al. paper; Pangolin primary paper https://doi.org/10.1186/s13059-022-02664-4, Methods: Deep neural network architecture, direct comparison with SpliceAIThe tool evaluates reference and alternative alleles and reports predicted acceptor/donor gains and losses with their relative positions. Annotation, search distance and masking determine the reported variant scores.
illumina/SpliceAI official source · README.md: package description, License, Usage and Frequently Asked QuestionsRelease 2026-09-16-d74d282221a9 · 0 evaluations · 0 metric rows. Different protocols are not a single leaderboard.
No evaluations linked in this release.
Primary project documentation or paper inspected for the explanatory claims and cited locations. Reviewed coverage concerns this narrative, not complete metadata, independent reproduction or a performance ranking.
Stable record: discovery-model-spliceaiThese evaluated pipelines include additional processing or trained components. Their results are not assigned to the underlying model.
Release 2026-09-16-d74d282221a9 · Record review: discovered
Stable ID: discovery-model-spliceai