Genomic Analysis of 11,555 Probands Provides a Framework for Congenital Heart Disease Genetics

Date of Award

Spring 2021

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Lifton, Richard


This dissertation presents the results from a genomic analysis of 11,555 congenital heart disease (CHD) probands with deep clinical phenotyping using a custom Molecular Inversion Probe Sequencing (MIPS) gene panel in order to further elucidate the genetic etiology of human CHD. CHD is a structural defect in the heart arising during development and afflicting ~1% of live newborns. CHD is a phenotypically complex and genetically heterogenous disease that frequently presents with congenital extracardiac abnormalities (EC) and neurodevelopmental deficits (NDD). The etiology for a majority of CHD cases is unknown and genotype-phenotype correlation is rare. Although hundreds of genes have been implicated in the pathogenesis of human CHD through prior investigation of multigenerational families with recurrent CHD and unrelated cohorts with sporadic CHD, few of these have been conclusively individually associated with human CHD. Herein, we designed and validated a low-cost, high-throughput MIPS panel of 248 known and candidate CHD genes. This is one of the largest MIPS panels described to-date. Furthermore, we describe a novel MIPS optimization strategy and pipeline that outperforms other panels of similar scope. We experimentally demonstrate this efficacy by performing MIPS on 6,069 CHD probands with high base coverage at a 98% success rate. We believe this panel can serve as a framework for interrogation of other complex diseases and, accordingly, describe effective application of this optimization strategy and pipeline to cerebral cavernous malformations (CCM). Subsequently, we aggregated cases with MIPS and whole exome sequencing together to generate a combined cohort of 11,555 CHD cases, including 3,869 proband-parent trios. This is the largest next generation sequencing cohort of severe structural CHD cases to-date. By performing a comprehensive genomic analysis on these cases, we implicate 61 genes in CHD pathogenesis with robust statistical confidence, including 14 not previously reported in human CHD. Collectively, we show protein-damaging mutations in these 61 genes account for 6.1% of CHD cases and show these genes alone recapitulate a majority (56%) of the disease-causing exome-wide de novo signal. Additionally, we show a distinct stratification between genes that contribute to CHD based on variant transmission modality and functional consequence, suggesting a distinct genetic architecture in CHD gene pathogenesis. We also performed detailed dissection of cardiac and extracardiac phenotypes to identify 44 genes significantly enriched in one or more cardiac lesions and 35 genes enriched in CHD cases with EC or NDD. We further interrogated genotype-phenotype relationships to reveal several novel associations, including that cysteine-related mutations in NOTCH1 EGF domains may disrupt Notch signaling and cause tetralogy of Fallot. We additionally compared the extracardiac manifestations between subjects with a molecular versus a clinical diagnosis of CHARGE, Kabuki, and Noonan syndromes and found no difference in clinically actionable phenotypes which suggests that early genomic evaluation of CHD patients to identify pathogenic syndromic variants can better inform patient care and clinical management versus clinical diagnosis alone. In conclusion, this dissertation presents a genomic investigation of 11,555 CHD probands using a novel MIPS gene panel that has expanded the current understanding of CHD genetics.

This document is currently not available here.