Integration of Large-Scale Genomic Data Sources with Evolutionary History Reveals Novel Genetic Loci for Congenital Heart Disease

Abstract

Background: Most cases of congenital heart disease (CHD) are sporadic and nonsyndromic, with poorly understood etiology. Rare genetic variants have been found to affect the risk of sporadic, nonsyndromic CHD, but individual studies to date are of only moderate sizes, and none to date has incorporated the ohnolog status of candidate genes in the analysis. Ohnologs are genes retained from ancestral whole-genome duplications during evolution; multiple lines of evidence suggest ohnologs are overrepresented among dosage-sensitive genes. We integrated large-scale data on rare variants with evolutionary information on ohnolog status to identify novel genetic loci predisposing to CHD. Methods: We compared copy number variants present in 4634 nonsyndromic CHD cases derived from publicly available data resources and the literature, and >27 000 healthy individuals. We analyzed deletions and duplications independently and identified copy number variant regions exclusive to cases. These data were integrated with whole-exome sequencing data from 829 sporadic, nonsyndromic patients with Tetralogy of Fallot. We placed our findings in an evolutionary context by comparing the proportion of vertebrate ohnologs in CHD cases and controls. Results: Novel genetic loci in CHD cases were significantly enriched for ohnologs compared with the genome (χ2 test, P<0.0001, OR =1.253 [95% CI, 1.199-1.309]). We identified 54 novel candidate protein-coding genes supported by both: (1) copy number variant and whole-exome sequencing data; and (2) ohnolog status. Conclusions: We have identified new CHD candidate loci, and show for the first time that ohnologs are overrepresented among CHD genes. Incorporation of evolutionary metrics may be useful in refining candidate genes emerging from large-scale genetic evaluations of CHD.