A validated heart-specific model for splice-disrupting variants in childhood heart disease

R. Lesurf; J. Breckpot; J. Bouwmeester; N. Hanafi; A. Jain; Y. Liang; T. Papaz; J. Lougheed; T. Mondal; M. Alsalehi; L. Altamirano-Diaz; E. Oechslin; E. Audain; G. Dombrowsky; A. V. Postma; O. I. Woudstra; B. J. Bouma; M.-P. Hitz; C. R. Bezzina; G. Blue; D. S. Winlaw; Seema Mital; Seema Mital

doi:https://doi.org/10.1101/2023.11.23.23298903

A validated heart-specific model for splice-disrupting variants in childhood heart disease

R. Lesurf, J. Breckpot, J. Bouwmeester, N. Hanafi, A. Jain, Y. Liang, T. Papaz, J. Lougheed, T. Mondal, M. Alsalehi, L. Altamirano-Diaz, E. Oechslin, E. Audain, G. Dombrowsky, A. V. Postma, O. I. Woudstra, B. J. Bouma, M.-P. Hitz, C. R. Bezzina, G. BlueD. S. Winlaw, Seema Mital, Seema Mital

Research output: Working paper › Preprint › Research

Abstract

Congenital heart disease (CHD) is the most common congenital anomaly. Non-canonical splice-disrupting variants are not routinely evaluated by clinical tests. Algorithms including SpliceAI predict such variants, but are not specific to cardiac-expressed genes. Whole genome (WGS) (n=1083) and myocardial RNA-Sequencing (RNA-Seq) (n=114) of CHD cases was used to identify splice-disrupting variants. Using features of variants confirmed to affect splicing in myocardial RNA, we trained a machine learning model that outperformed SpliceAI for predicting cardiac-specific splice-disrupting variants (AUC 0.92 vs 0.66), and was independently validated in 43 cardiomyopathy probands (AUC 0.88 vs 0.64). Application of this model to 971 CHD WGS samples identified 9% patients with splice-disrupting variants in CHD genes. Forty-one% of predicted splice-disrupting variants were deeply intronic. The burden of variants in CHD genes was higher in cases compared with 2,570 controls. Our model improved genetic yield by identifying splice-disrupting variants that are not evaluated by routine tests.

Original language	English
DOIs	https://doi.org/10.1101/2023.11.23.23298903
Publication status	Published - 2023

Keywords

genetic and genomic medicine

Access to Document

https://doi.org/10.1101/2023.11.23.23298903Licence: CC BY-NC-ND

Cite this

Lesurf, R., Breckpot, J., Bouwmeester, J., Hanafi, N., Jain, A., Liang, Y., Papaz, T., Lougheed, J., Mondal, T., Alsalehi, M., Altamirano-Diaz, L., Oechslin, E., Audain, E., Dombrowsky, G., Postma, A. V., Woudstra, O. I., Bouma, B. J., Hitz, M.-P., Bezzina, C. R., ... Mital, S. (2023). A validated heart-specific model for splice-disrupting variants in childhood heart disease. https://doi.org/10.1101/2023.11.23.23298903

@techreport{bf68df3c6e3a40a0a8e8e4f6f1107456,

title = "A validated heart-specific model for splice-disrupting variants in childhood heart disease",

abstract = "Congenital heart disease (CHD) is the most common congenital anomaly. Non-canonical splice-disrupting variants are not routinely evaluated by clinical tests. Algorithms including SpliceAI predict such variants, but are not specific to cardiac-expressed genes. Whole genome (WGS) (n=1083) and myocardial RNA-Sequencing (RNA-Seq) (n=114) of CHD cases was used to identify splice-disrupting variants. Using features of variants confirmed to affect splicing in myocardial RNA, we trained a machine learning model that outperformed SpliceAI for predicting cardiac-specific splice-disrupting variants (AUC 0.92 vs 0.66), and was independently validated in 43 cardiomyopathy probands (AUC 0.88 vs 0.64). Application of this model to 971 CHD WGS samples identified 9% patients with splice-disrupting variants in CHD genes. Forty-one% of predicted splice-disrupting variants were deeply intronic. The burden of variants in CHD genes was higher in cases compared with 2,570 controls. Our model improved genetic yield by identifying splice-disrupting variants that are not evaluated by routine tests.",

keywords = "genetic and genomic medicine",

author = "R. Lesurf and J. Breckpot and J. Bouwmeester and N. Hanafi and A. Jain and Y. Liang and T. Papaz and J. Lougheed and T. Mondal and M. Alsalehi and L. Altamirano-Diaz and E. Oechslin and E. Audain and G. Dombrowsky and Postma, {A. V.} and Woudstra, {O. I.} and Bouma, {B. J.} and M.-P. Hitz and Bezzina, {C. R.} and G. Blue and Winlaw, {D. S.} and Seema Mital and Seema Mital",

year = "2023",

doi = "https://doi.org/10.1101/2023.11.23.23298903",

language = "English",

type = "WorkingPaper",

}

Lesurf, R, Breckpot, J, Bouwmeester, J, Hanafi, N, Jain, A, Liang, Y, Papaz, T, Lougheed, J, Mondal, T, Alsalehi, M, Altamirano-Diaz, L, Oechslin, E, Audain, E, Dombrowsky, G, Postma, AV, Woudstra, OI, Bouma, BJ, Hitz, M-P, Bezzina, CR, Blue, G, Winlaw, DS, Mital, S & Mital, S 2023 'A validated heart-specific model for splice-disrupting variants in childhood heart disease'. https://doi.org/10.1101/2023.11.23.23298903

TY - UNPB

T1 - A validated heart-specific model for splice-disrupting variants in childhood heart disease

AU - Lesurf, R.

AU - Breckpot, J.

AU - Bouwmeester, J.

AU - Hanafi, N.

AU - Jain, A.

AU - Liang, Y.

AU - Papaz, T.

AU - Lougheed, J.

AU - Mondal, T.

AU - Alsalehi, M.

AU - Altamirano-Diaz, L.

AU - Oechslin, E.

AU - Audain, E.

AU - Dombrowsky, G.

AU - Postma, A. V.

AU - Woudstra, O. I.

AU - Bouma, B. J.

AU - Hitz, M.-P.

AU - Bezzina, C. R.

AU - Blue, G.

AU - Winlaw, D. S.

AU - Mital, Seema

PY - 2023

Y1 - 2023

N2 - Congenital heart disease (CHD) is the most common congenital anomaly. Non-canonical splice-disrupting variants are not routinely evaluated by clinical tests. Algorithms including SpliceAI predict such variants, but are not specific to cardiac-expressed genes. Whole genome (WGS) (n=1083) and myocardial RNA-Sequencing (RNA-Seq) (n=114) of CHD cases was used to identify splice-disrupting variants. Using features of variants confirmed to affect splicing in myocardial RNA, we trained a machine learning model that outperformed SpliceAI for predicting cardiac-specific splice-disrupting variants (AUC 0.92 vs 0.66), and was independently validated in 43 cardiomyopathy probands (AUC 0.88 vs 0.64). Application of this model to 971 CHD WGS samples identified 9% patients with splice-disrupting variants in CHD genes. Forty-one% of predicted splice-disrupting variants were deeply intronic. The burden of variants in CHD genes was higher in cases compared with 2,570 controls. Our model improved genetic yield by identifying splice-disrupting variants that are not evaluated by routine tests.

AB - Congenital heart disease (CHD) is the most common congenital anomaly. Non-canonical splice-disrupting variants are not routinely evaluated by clinical tests. Algorithms including SpliceAI predict such variants, but are not specific to cardiac-expressed genes. Whole genome (WGS) (n=1083) and myocardial RNA-Sequencing (RNA-Seq) (n=114) of CHD cases was used to identify splice-disrupting variants. Using features of variants confirmed to affect splicing in myocardial RNA, we trained a machine learning model that outperformed SpliceAI for predicting cardiac-specific splice-disrupting variants (AUC 0.92 vs 0.66), and was independently validated in 43 cardiomyopathy probands (AUC 0.88 vs 0.64). Application of this model to 971 CHD WGS samples identified 9% patients with splice-disrupting variants in CHD genes. Forty-one% of predicted splice-disrupting variants were deeply intronic. The burden of variants in CHD genes was higher in cases compared with 2,570 controls. Our model improved genetic yield by identifying splice-disrupting variants that are not evaluated by routine tests.

KW - genetic and genomic medicine

U2 - https://doi.org/10.1101/2023.11.23.23298903

DO - https://doi.org/10.1101/2023.11.23.23298903

M3 - Preprint

BT - A validated heart-specific model for splice-disrupting variants in childhood heart disease

ER -