USING DISCORDANT SIB PAIRS TO MAP LOCI FOR QUALITATIVE TRAITS WITH HIGH SIBLING RECURRENCE RISK

A common approach for detecting genetic linkage using siblings is to c ollect affected sib pairs (ASPs) and to identify markers where allele sharing exceeds expectation. Alternatively, markers can be analyzed in discordant sih pairs (DSPs) for allele sharing below expectation. Rel ative to the ASP approach, according to Risch, the power of the DSP ap proach increases with sibling recurrence risk, the two approaches bein g equally effective at 50% recurrence risk. However, with many disease s associated with more moderate sibling recurrence risk, less emphasis has been placed on the use of DSPs and the development of the underly ing theory. In this paper, we expand the work of Risch to provide a mo re general foundation for DSP studies, since power can be quite high u nder the appropriate conditions. For example, in some highly affected populations, such as the diabetes-prone Pima Indians, sibling recurren ce risk can be very large and, thus, DSPs ideal. Similarly, as we show through simulation, DSPs are preferable for diabetic nephropathy due to a 70% recurrence rare among siblings with insulin-dependent diabete s mellitus. Following the diabetic nephropathy example, we consider mo re systematically the situations in which DSPs can provide an efficien t alternative to ASPs.