IMPACT OF ADJUSTMENTS FOR INTERMEDIATE PHENOTYPES ON THE POWER TO DETECT LINKAGE

Since the manifestation of a complex disease is likely to be influence d through multiple genetic and/or environmental pathways, it may be ad vantageous to adjust for these multiple factors in a genetic analysis of a complex quantitative trait. Sib-pair linkage analysis was perform ed on the simulated complex quantitative trait Q1 after adjustment for age, sex, and the environmental factor (i.e., minimally adjusted) and all combinations of the four intermediate phenotypes Q2, Q3, Q4, and Q5 (n = 15) for all 200 replications of the nuclear families data set. From the minimally adjusted Q1, the power to detect suggestive linkag e to any of the three loci affecting Q1 was 0.585 with a false positiv e rate of 0.0025. Adjusting Q1 for Q3 increased the power to detect su ggestive linkage to 0.860 with a similar false positive rate. Addition al adjustments for Q2, Q4, and Q5 yielded no substantial improvements in power nor changes in the false positive rate. The power to detect s ignificant linkage was also substantially improved after adjustment of Q1 for Q3 with no change in the false positive rate. The adjustment o f a complex trait for other factors in the causal pathway reduces the phenotype variability and enhances the ability to detect linkage. (C) 1997 Wiley-Liss, Inc.