Combining extremely concordant sibpairs with extremely discordant sibpairs provides a cost effective way to linkage analysis of quantitative trait loci

Extremely discordant (ED) sibpairs have been shown to be very powerful for linkage analysis of human quantitative traits [Risch and Zhang (1995) Science 268:1584-1589]. In many cases, the extremely concordant (EC) sibpairs collected in the process of screening for ED sibpairs carry valuable information for linkage. Therefore, it seems justifiable to investigate the advantages of genotyping and to include them with the ED sibpairs for linkage analysis. Herein we explore the distributions of EC as well as ED sibpairs under various genetic models and provide a basis for combining both types of sibpairs. A simple statistic testing means of genes shared identical by descent (IBD) is applied to combine both types of EC sibpairs (high-high and low-low) with the ED pairs. We show that when a decent number of EC pairs is added to the ED sample for analysis, the power is much enhanced, making it especially desirable when the number of available ED pairs is small. We show at the same time that combining EC pairs with ED pairs is more cost effective than pursuing ED sibpairs alone.