Power of the linkage test for a heterogeneous disorder due to two independent inherited causes: a simulation study.

We have conducted a simulation study in small pedigrees to investigate the power to detect linkage and heterogeneity for a disorder due to either one of two independent disease loci. We have considered a highly polymorphic marker locus (PIC = 70%) linked to one disease locus and unlinked to the second. The power to detect linkage has been examined by using the admixture test. We have varied the mode of transmission of each disease locus, the ascertainment of families and the proportion of cases in the population due to the linked disease locus. Generally, for the multiplex ascertainments we have considered, the power to detect linkage is greater when the linked disease locus has a high penetrance, when the unlinked disease locus has a low penetrance, and when pedigrees with multiple affected are selected. When selecting families with multiple affected, the rate of "mixed" families (i.e., those segregating for both disease loci) increases. However, for the pedigree structure we have considered, the power of the linkage test is more affected by a decrease in the rate of "linked" families than by an increase in the rate of "mixed" families. Finally, the present study shows that detection of linkage in presence of heterogeneity is feasible with a realistic sample size of small pedigrees as long as the linked disease locus accounts for more than 25% of the cases.