A STUDY OF SEGREGATION/LINKAGE ANALYSIS FOR HYPERTENSION

本项目拟继续进行一系列模拟实验

This project proposes to continue a series of simulation experiments designed to determine the power, robustness and validity of recently developed methods of segregation and linkage analysis. Special emphasis will be placed on determining the power and robustness of methods of segregation analysis that incorporate corrections for certainment bias and to determine the effect, if any, of failing to correct for the method of ascertainment, or of using an inappropriate correction. Insight gained from these experiments will be used to analyze pedigree data on blood pressure and other variables related to hypertension. A particular effort will be made to detect major genes in an attempt to determine the genetic component of hypertension. The power, robustness and validity of three genetic models will be considered: 1) the transmission probability model, 2) the mixed model, and 3) the unified model, a mixed model with the single locus component parameterized in terms of transmission probabilities. The robustness of each model will be considered with respect to the presence of environmental skewness and kurtosis, and other familial environmental correlations. Several methods of correcting for ascertainment bias will be used in the analysis of data selected according to various sampling schemes (random, sequential, proband sampling frame, "crystal growth", and simplex and multiplex ascertainment). Simulated data will be based on parameters derived from actual hypertension studies, so that valid conclusions can be drawm from these experiments. Methods of segregation and linkage analysis that are powerful and robust with respect to failures of assumptions (particularly those found in the actual hypertension data) will be applied to two sets of data provided by other investigators. The primary set of data comprises five large pedigrees each ascertained through a single hypertensive proband. The second set of data will consist of approximately ten large pedigrees ascertained at random as part of the Bogalusa Heart Study. Long range objectives include 1) the dissection of the genetic component of hypertension with a view to the identification of individuals with a genetic predisposition for hypertension before those individuals become overtly hypertensive and 2) an ongoing evaluation of new methods in segregation/linkage analysis, used to illuminate the nature of genetic components.