Exome-based linkage mapping and variant prioritization for inherited retinal disorders. D. C. Koboldt1, D. E. Larson1, L. S. Sullivan2, S. J. Bowne2, R. S. Fulton1, E. Sodergren1, S. H. Blanton3, K. Meltz Steinberg1, S. P. Daiger2, R. K. Wilson1, G. W. Weinstock1 1) The Genome Institute at Washington University, St. Louis, MO; 2) Human Genetics Center, Univ. of Texas Health Science Ctr, Houston, TX; 3) Hussman Institute of Human Genomics, Univ. of Miami, Miami, FL.

   Exome sequencing in families with rare genetic disorders has the potential to rapidly identify new disease genes. Yet individual exomes harbor thousands of coding variants, and the identification of a single causal mutation among them remains a significant challenge. We developed a scoring algorithm to prioritize potential causal variants within a family according to segregation with the phenotype, population frequency, predicted effect, and gene expression in the tissue(s) of interest. We also describe two complementary approaches to exome-based linkage analysis that help narrow the search space in families with multiple affected individuals: (1) Shared IBD analysis of common variants identifies segments of maximum identity-by-descent among affected individuals; (2) Rare heterozygote rule out nominates regions based on shared rare variants and the absence of homozygous differences between affected individuals. We showcase our methods using exome sequence data from families with autosomal dominant retinitis pigmentosa (adRP), a rare disorder characterized by progressive vision loss. We performed exome capture and sequencing on 75 samples representing 24 families with probable adRP but lacking common disease-causing mutations. A subset of these families also had regions from traditional linkage mapping of extended pedigrees, the results of which were highly concordant with our linkage analyses. Seven of 24 families (29.17%) were revealed to have unrecognized mutations in known RP genes that were both high-scoring by our scoring algorithm and deemed likely pathogenic by clinical assessment. Analysis of the remaining 17 families has identified candidate variants in a number of interesting genes, some of which have withstood further segregation testing in extended pedigrees. Family-based exome sequencing is a powerful strategy for the identification of novel disease genes, yet these studies often identify thousands of potential causative variants. Here, we demonstrate that comprehensive scoring of individual variants coupled with two genetic linkage approaches can substantially refine the search for disease-causing mutations.

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