Introme accurately predicts the impact of coding and noncoding variants on gene splicing, with clinical applications

Patricia J Sullivan; Velimir Gayevskiy; Ryan L Davis; Marie Wong; Chelsea Mayoh; Amali Mallawaarachchi; Yvonne Hort; Mark J McCabe; Sarah Beecroft; Matilda R Jackson; Peer Arts; Andrew Dubowsky; Nigel Laing; Marcel E Dinger; Hamish S Scott; Emily Oates; Mark Pinese; Mark J Cowley

doi:10.1186/s13059-023-02936-7

Introme accurately predicts the impact of coding and noncoding variants on gene splicing, with clinical applications

Genome Biol. 2023 May 17;24(1):118. doi: 10.1186/s13059-023-02936-7.

Authors

Patricia J Sullivan^{1

2

3}, Velimir Gayevskiy⁴, Ryan L Davis^{4

5

6}, Marie Wong^{1

2}, Chelsea Mayoh^{1

2}, Amali Mallawaarachchi^{7

8}, Yvonne Hort⁷, Mark J McCabe⁴, Sarah Beecroft⁹, Matilda R Jackson^{10

11}, Peer Arts¹⁰, Andrew Dubowsky¹², Nigel Laing⁹, Marcel E Dinger^{4

13}, Hamish S Scott^{10

11

14

15}, Emily Oates¹³, Mark Pinese^{1

2}, Mark J Cowley^{16

17}

Affiliations

¹ Children's Cancer Institute, Lowy Cancer Research Centre, UNSW Sydney, Sydney, NSW, Australia.
² School of Clinical Medicine, UNSW Medicine & Health, UNSW Sydney, Sydney, NSW, Australia.
³ University of New South Wales Centre for Childhood Cancer Research, UNSW Sydney, Sydney, NSW, Australia.
⁴ Kinghorn Centre for Clinical Genomics, Garvan Institute of Medical Research, Sydney, Australia.
⁵ Department of Neurogenetics, Kolling Institute, St. Leonards, NSW, Australia.
⁶ Sydney Medical School-Northern, Faculty of Medicine and Health, University of Sydney, Sydney, NSW, Australia.
⁷ Division of Genomics and Epigenetics, Garvan Institute of Medical Research, Sydney, Australia.
⁸ Clinical Genetics Unit, Institute of Precision Medicine and Bioinformatics, Sydney Local Health District, Sydney, Australia.
⁹ Centre for Medical Research, University of Western Australia, Harry Perkins Institute of Medical Research, QEII Medical Centre, Nedlands, WA, Australia.
¹⁰ Department of Genetics and Molecular Pathology, Centre for Cancer Biology, An Alliance Between SA Pathology and the University of South Australia, Adelaide, Australia.
¹¹ Australian Genomics, Parkville, VIC, Australia.
¹² Department of Genetics and Molecular Pathology, SA Pathology, Adelaide, Australia.
¹³ School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia.
¹⁴ School of Medicine, University of Adelaide, Adelaide, SA, Australia.
¹⁵ ACRF Cancer Genomics Facility, Centre for Cancer Biology, An Alliance Between SA Pathology and the University of South Australia, Adelaide, SA, Australia.
¹⁶ Children's Cancer Institute, Lowy Cancer Research Centre, UNSW Sydney, Sydney, NSW, Australia. MCowley@ccia.org.au.
¹⁷ School of Clinical Medicine, UNSW Medicine & Health, UNSW Sydney, Sydney, NSW, Australia. MCowley@ccia.org.au.

Abstract

Predicting the impact of coding and noncoding variants on splicing is challenging, particularly in non-canonical splice sites, leading to missed diagnoses in patients. Existing splice prediction tools are complementary but knowing which to use for each splicing context remains difficult. Here, we describe Introme, which uses machine learning to integrate predictions from several splice detection tools, additional splicing rules, and gene architecture features to comprehensively evaluate the likelihood of a variant impacting splicing. Through extensive benchmarking across 21,000 splice-altering variants, Introme outperformed all tools (auPRC: 0.98) for the detection of clinically significant splice variants. Introme is available at https://github.com/CCICB/introme .

Keywords: Clinical genetics; Deep intronic; Genomics; Intronic variant; Splice region; Splice site; Splicing; Splicing regulatory element; Variant interpretation.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Humans
Introns
Machine Learning
Mutation
RNA Splice Sites*
RNA Splicing*

Substances

RNA Splice Sites