Genome-wide association study on secundum atrial septal defects. L. Rodriguez-Murillo1, M. Parfenov5, I. Peter2, W. K. Chung4, L. Mitchell3, A. J. Agopian3, C. Seidman5, J. Seidman5, B. D. Gelb1, Pediatric Cardiac Genomics Consortium 1) Mindich Child Health and Development Institute, Departments of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, NY; 2) Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, NY; 3) Human Genetics Center, Division of Epidemiology, Human Genetics and Environmental Sciences, University of Texas School of Public Health, Houston, TX; 4) Departments of Pediatrics and Medicine, Columbia University, New York, NY; 5) Department of Genetics, Harvard Medical School, Boston, MA.

   Secundum atrial septal defects (ASDs) are among the commonest forms of congenital heart disease (CHD), characterized by deficiencies in the septum primum that contributes to separation of the atria. To elucidate the contribution of common genetic variants to the etiology of ASD, the Pediatric Cardiac Genetics Consortium (PCGC) performed a genome-wide association study (GWAS) using a discovery cohort of 296 subjects with isolated ASD, genotyped with Illumina Omni2.5 microarrays. Our control cohort comprised two dbGaP datasets (916 individuals from a smoking cessation study genotyped with the Illumina Omni2.5 microarray and 2416 from the Framingham heart study genotyped with the Illumina Omni5 microarray). After quality control and principal component analysis (PCA) to correct for population stratification, 1,437,767 SNPs were tested for association on 165 ASD cases and 3330 controls (120 ASD cases showed mixed ethnicity after PCA, therefore were not included in the analysis). The strongest association in cases versus controls attained genome-wide significance using an additive model at SNP kgp10664470 (P=3x10-10; OR = 3.3). ~18 of our ASD cases harbor the minor allele for this SNP. SNPs in linkage disequilibrium with it also showed genome-wide significant P values (P<1.4x10-8), providing robustness to the signal. Subsequently, we replicated this finding by genotyping 144 additional PCGC ASD cases and testing for association against an independent dbGaP control dataset from an eye study (101 ASD cases and 1607 controls were included in the analysis after PCA correction for population stratification). All significant SNPs in the discovery dataset replicated our original findings with a consistent direction of association and a top combined P value of 7.8x10-11. These SNPs are located within the ROBO2 gene, which encodes an immunoglobulin superfamily protein that is evolutionarily conserved and has roles in cardiac cell polarity and morphogenesis in mice, zebrafish and Drosophila. This is a novel gene that had not been identified in previous GWAS performed either in Chinese population (Hu et al. 2013) or Caucasian population (Cordell et al. 2013). Confirmation in larger ASD cohorts and assessment of whether ROBO2 genetic variation contributes to ASD is underway.

You may contact the first author (during and after the meeting) at