Theses Doctoral

Mosaicism and the genetic architecture of congenital heart disease

Hsieh, Alexander Lin

Congenital heart disease (CHD) is characterized by structural defects of the heart and great vessels. It is the most common birth defect, affecting an estimated 1% of live births, and is the leading cause of mortality among birth defects. Despite recent progress in genetic research, more than 50% of CHD cases remain unexplained. An estimated 23% are due to aneuploidies and copy number variants and up to 30% has been attributed to de novo variation, though that number ranges between 3-30% depending on CHD complexity.

The contribution of somatic mosaicism, or de novo genetic mutations arising after oocyte fertilization, to congenital heart disease (CHD) is not well understood due to limitations in sample size, detection method, and validation rate. Further, the relationship between mosaicism in blood and cardiovascular tissue has not been determined. We developed a computational method, Expectation-Maximization-based detection of Mosaicism (EM-mosaic), to analyze mosaicism in exome sequences of 2530 CHD proband-parent trios. EM-mosaic accurately detected 309 mosaic mutations in blood, with 85 of 94 (90%) candidates tested independently confirmed. We found twenty-five likely damaging mosaics in plausible CHD-risk genes, affecting 1% of our cohort. Variants in these genes predicted as damaging had higher variant allele fraction than benign variants, suggesting a role in CHD. The frequency of protein-coding mosaic variants detectable in blood was 0.122 or roughly 1 in 8 individuals. Analysis of 66 individuals with matched cardiac tissue available revealed both tissue-specific and shared mosaicism, with shared mosaics generally having higher allele fraction.

CHD patients often present with comorbid cardiac and extracardiac anomalies that further their impact quality of life. Neurodevelopmental disorders (NDDs) are especially prevalent in CHD cases compared to the general population, yet the underlying genetic causes remain poorly explained. Further, patients with single ventricle defects undergoing surgery often later develop arrhythmias and experience worsening ventricular function. We used a statistical approach to dissect the association between de novo variation and these clinical outcomes and found that pleiotropic mutations contribute a large fraction of the risk of acquiring NDD and abnormal ventricular function phenotypes in CHD patients. We developed a proof-of-concept rare variant risk score that combines information from de novo, rare transmitted, and copy- number variants and show that prediction of outcomes such as NDD can be improved, especially in complex CHD cases.


  • thumnail for supp_chd_mosaicism_tables_9Aug2019.xlsx supp_chd_mosaicism_tables_9Aug2019.xlsx application/vnd.openxmlformats-officedocument.spreadsheetml.sheet 1.62 MB Download File
  • thumnail for Hsieh_columbia_0054D_15672.pdf Hsieh_columbia_0054D_15672.pdf application/pdf 2.57 MB Download File

More About This Work

Academic Units
Biomedical Informatics
Thesis Advisors
Shen, Yufeng
Ph.D., Columbia University
Published Here
January 23, 2020