EFFECTS OF GENOTYPE-BY-SEX INTERACTION ON QUANTITATIVE TRAIT LINKAGE ANALYSIS

Genotype-by-sex (G x S) interaction refers to the interaction of autos omal genes with male or female physiological ''environments.'' G x S i nteraction has been demonstrated in quantitative genetic analyses of a variety of traits including serum lipid concentrations and anthropome trics, and the importance of considering sex-specific major gene effec ts in segregation analyses also has been demonstrated. The goal of thi s study was to examine the effects of G x S interaction on the power t o detect linkage. Trait Q3 in GAW10 Problem 2 was analyzed because it was modeled to have G x S interaction at the major gene locus MG3. All 200 nuclear family and 200 extended pedigree replicates were first sc reened for the presence of G x S interaction in Q3 using a quantitativ e genetic method. More than half of both the nuclear family and extend ed pedigree replicates evidenced significant G x S interaction. Varian ce components linkage analysis was then performed using all markers on GAW10 chromosome 4 in all 200 nuclear family and 200 extended pedigre e replicates. A peak lod score of 1.92 at the correct chromosomal loca tion was obtained using the extended pedigree data and incorporating G x S interaction effects. Not incorporating G x S interaction lowered the peak lod score from the analyses of the extended pedigrees to 1.53 . Incorporation of G x S interaction effects also increased the power to detect linkage in the nuclear family replicates, although the nucle ar families had considerably less power than the extended pedigrees to detect linkage, whether or not G x S interaction was modeled. Incorpo ration of G x S interaction effects can increase the power to detect l inkage, even when the G x S interaction effects are modest. (C) 1997 W iley-Liss, Inc.