# Notices and attribution

## SpliceAI

The scores were produced using SpliceAI 1.3.1 and its five-model ensemble.
SpliceAI was developed by Illumina, Inc. This dataset is independent and is not
affiliated with or endorsed by Illumina. The SpliceAI name is used only to
identify compatibility with the upstream method and score format.

Please cite:

Jaganathan K, et al. Predicting Splicing from Primary Sequence with Deep
Learning. *Cell*. 2019;176(3):535–548.e24.
<https://doi.org/10.1016/j.cell.2018.12.015>

## MANE

Transcript definitions are derived from MANE Select v1.5. The source MANE
annotation itself is not redistributed in this dataset.

Please cite:

Morales J, et al. A joint NCBI and EMBL-EBI transcript set for clinical
genomics and research. *Nature*. 2022;604:310–315.
<https://doi.org/10.1038/s41586-022-04558-8>

## Reference genome

Scores use GRCh38/hg38 coordinates. Reference genome sequence is not included.

## Generation software

The generation software is available separately under GPL-3.0-or-later:

<https://github.com/yimingluo-md/batched-inference-for-spliceai>

The dataset license does not alter the software license, and the software
license does not replace the dataset license.

## External validation

The Broad SpliceAI Lookup service was used as an external scientific
comparator. Raw API responses are not included. Broad Institute does not
endorse this dataset.

## Research-use notice

This dataset is not a validated clinical diagnostic device. Predictions require
independent review and must not be the sole basis for patient-care decisions.