Using family history information to distinguish true and false positive model-free linkage results.

Genome scans that test for increased marker identity-by-descent sharing between pairs of affected siblings have become increasingly common. These methods do not specify a priori a genetic model for the disease locus and as such lose the ability to specify the parental source of the disease allele. We propose a method that uses family history information to build a more complete model of disease and marker inheritance, while still avoiding specification of the parameters of the disease model of inheritance. One important use for such a model is to test whether a positive linkage result obtained during the course of a genome scan is a true or false positive result. The key to the new test statistics is the interaction between gender-specific marker identity-by-descent sharing and gender-specific family history of disease. The method is useful when the disease locus of interest has a dominant mode of inheritance and a sufficient number of parents are genotyped at the marker locus. If these conditions are met, the proposed tests have good power to differentiate between true and false positive linkage results.