Abstract 2726: Haplotype Sharing Test as a Tool to Map Genes for Familial Cardiomyopathy

Abstract

Background: In Mendelian diseases, such as arrhythmogenic right ventricular and dilated cardiomyopathies (ARVC and DCM), chromosomal regions identical-by-descent (IBD) from a common founder can be ascertained both by pedigree analysis and by haplotype-sharing methods. Finding shared haplotypes is greatly facilitated with the currently available high-density SNP arrays. However, determining which of the shared haplotypes is IBD and contains the disease-associated mutation, and which are identical-by-state (IBS) and are shared by chance, is difficult. An additional criterion is needed to indicate the region containing the disease mutation. The probability for shared haplotypes to be IBD rather than IBS increases with an increasing number of SNPs. We hypothesized that the largest shared haplotype is the most likely region to hold the causative disease mutation. Methods: The Haplotype-Sharing Test (HST) was designed, using SNP genotyping data from isolated patients and parent-offspring pairs and trios to identify the largest possibly shared haplotypes between patients that are members of a (large but unobserved) pedigree. We applied the HST to: three distantly related families each with at least one ARVC patient using 10K SNP genotyping arrays, and a large family with 5 DCM patients using 250K SNP arrays. Results: In pedigree A a haplotype run of 118 SNPs spanning 32 MB on chromosome 12p12.3-q13.13 was identified. This haplotype is substantially larger than any other area that is shared due to random effects. Screening of the PKP2 gene, located in this region, revealed a pathogenic splice-site mutation that was segregating with the disease. In family B (DCM) the largest shared haplotype was 178 SNPs, spanning 3.5 MB on chromosome 15. Sequencing of potential candidate genes (including alpha Tropomyosin and Integrin alpha 1) is pending. Conclusions: Our identification of the causative founder mutation in pedigree A in the largest shared region shows that our hypothesis, though heuristic in character, was correct. More importantly, haplotype sharing is a powerful tool for identifying disease-causing genes in single and extended families when an insufficient number of meioses are present to perform classic linkage analysis.