For the family-based replication study, SNP association analysis was conducted with the statistical computer software SOLAR [Almasy and Blangero, 1998] using the quantitative trait linkage disequilibrium (QTLD) procedure. This procedure performs a test for population stratification (based on the likelihood comparison of between-family and within-family association parameters of estimated and constrained fixed effect models) and two commonly used association tests: measured genotype test (see above) and the quantitative transmission disequilibrium test (QTDT) [Abecasis et al., 2000]. Although QTDT is a family-based association test that controls for spurious associations due to population stratification by scoring allelic transmissions based on parental genotypes, in the absence of stratification the classical measured genotype approach uses more of the relative association information and exhibits more power than the QTDT [Havill et al., 2005] and thus is reported for the replication results when this condition is met.
