Abstract 16717: Human Gene Copy Number Spectra Analysis in Congenital Heart Malformations

Aim: To visualize copy number associations and quantify relative risk of Copy Number Variants (CNVs) in 100 well-defined congenital heart disease (CHD) risk genes identified from literature sources. Method: CNVs in subjects with congenital heart malformations (n=741) and phenotyped in accordance with the European Paediatric Cardiac Code were compared to the frequencies of CNVs present in a control population of subjects with coronary artery disease (n=880). CNV gains (≥ 200 kb) and losses (≥ 100 kb) were determined using the Affymetrix 6.0 Gene Chip and confirmed with TaqMan copy number assays. In addition, each gene was tested using a two proportions z-test approach to determine if the gene frequency was significantly different in CHD vs control. To quantify the effect of observed CNVs in the CHD population, we collapsed the set of rare individual CNVs into a single group to test the collective frequency difference between cases and controls by adapting a previously described Cohort Allelic Sums Test (CAST). Results: The following phenotypes showed a significant enrichment of gains and losses in the CHD population: Aortic Stenosis, Coarctation of the Aorta, Ebstein's Anomaly, Tetralogy of Fallot, Truncus Arteriosus and Ventricular Septal Defect. The CNV percentage was increased in CHD subjects in the following genes (p<0.05 with a Bonferroni correction): Losses: FKBP6(-), ELN(s), GTF2IRD1(-), SOX7(-), GATA4(ns), NOTCH1(ns), MYH11(ns), SALL4(s), TBX1(s), CRKL(-), MAPK1(-); Gains: PRKAB2(-), FMO5(-), CHD1L(-), BCL9(-), ACP6(-), GJA5(ns), NPHP3(s), FOXL2(s), SEMA5A(-), NSD1(s), HOXA1(s), TBX20(ns), SOX7(-), GATA4(ns), HRAS(s), GATA6(ns), RUNX1(s), MAPK1(-), BCOR(s), ATRX(s), GPC3(s), ZIC3(ns), FLNA(s) where s is syndromic, ns is non-syndromic and (-) is not available on the CHD WIKI website ( http://www.chearted.eu ). The approach rapidly identified common, cytogenetically visible abnormalities (i.e. TBX1 in 22q11.2 deletion syndrome, n=30, and RUNX1 in Trisomy 21, n=66). Conclusion: This is the first study to establish a copy number spectrum in a CHD population. We conclude that the CAST approach shows promise for evaluating qualitative outcomes and statistical precision of prior studies which have focused on CNVs in CHD and other diseases.