The long-term objective of this research program is to develop, evaluate, and apply valid, efficient, and timely methodologies for investigating the interplay between genes and environments in the etiology of complex multifactorial diseases and disease-related traits. Towards this goal, the specific aims cover theoretical/methodological studies in path, segregation, and linkage analysis, analysis of numerous actual family studies on diseases and risk factors, and simulation studies on timely methodological issues. The proposed theoretical-methodological studies include extensions of multivariate path analysis to incorporate multivariate models of environmental sources of familial resemblance using a new computer-based algorithm that enables a flexible approach to modeling; extensions of temporal trend methodology for investigation of temporal trends in familial aggregation of risk factors using direct effects of relevant covariates (e.g., the direct effect of obesity on blood pressure); combined path, segregation, and linkage analysis for resolution of the effects of familial environment, polygenes, and multiple measured and unmeasured major loci, incorporating linked markers. The combined models are especially motivated for complex diseases like chronic obstructive pulmonary disease and coronary artery disease. Analysis of numerous actual family studies conducted in diverse populations are proposed to investigate the determinants of familial aggregation for a variety of multifactorial traits, to evaluate the applicability and practical utility of the methods, and to identify specific problem areas that need further methodological work. Extensive and often challenging analyses of actual data include obesity and regional fat distribution, blood pressure, pulmonary function, lipids and lipoproteins, hemochromatosis and biochemical measures of iron metabolism, immunoglobulins and atopy, glucose tolerance and insulin concentrations, among others. Several simulation studies are proposed to evaluate the potential merits and pitfalls associated with various theoretical models, and to evaluate some important methodological issues that do not readily lend themselves to theoretical investigations. The complex nature of the state-of-the-art models makes it necessary and the electronics revolutions makes it possible to undertake the proposed studies on the practical utility of transmission probabilities in segregation analysis, identifying segregating families for linkage studies, and the power of path analysis, among others. Maximum likelihood method of estimation and likelihood ratio tests of hypotheses will be used to elucidate the univariate and multivariate interrelationships, and population heterogeneity. With emphasis on testing and generation of hypotheses, it is believed that the proposed multifaceted approach using theoretical/methodological studies, analysis of actual family studies, and simulation studies represents a significant advancement in the field of genetic epidemiology.
Showing the most recent 10 out of 160 publications
