An investigation of the genetic and environmental architecture of complex phenotypes, through an evaluation of a series of hypotheses pertaining to the interrelationships among the variables, constitutes the long-term goal of this (re-revised) proposal. This involves the development and evaluation of novel methods for evaluating the underlying hypotheses. Towards this end, the specific aims cover methodological issues in linkage and association analysis, dense SNP scans and haplotype methods, novel methods for taking advantage of gene-gene (GxG) and gene-environment (GxE) interactions, and novel sequential methods for bypassing the issue of multiple comparisons. Summer workshops will be held which include presentations on the methods and """"""""products"""""""" we develop(ed), hands-on computer lab sessions on the use of the products, and brain-storming sessions on the merits and pitfalls of these and other methods/products. All software developed will be freely distributed. The proposed research includes the development and evaluation of statistical methods for the analysis of phenotypic and genotypic data: timely and promising extensions of the variance components linkage models in SEGPATH to incorporate temporal trends in linkage analysis (to capture age-dependent genetic effects), parent-of-origin and imprinting effects, and LD and TDT-like methods; an assessment of the promise of dense SNP scans in candidate genes along with the analysts issues associated with them, theoretical challenges posed by the construction of haplotype maps, ambiguity of block boundaries arising on account of the ambiguity about the makeup of haplotype blocks, and the role of HapMap in mapping complex disease genes; novel methods for finding genes in the presence of GxG and GxE interactions in terms of """"""""tree linkage""""""""; and novel sequential methods for analysis of genome-wide linkage scans and association scans, with particular emphasis on tight control of false positives. Our direct access to several large collaborative family data sets, which have been phenotypically and genotypically well characterized, offers ample opportunity for testing the real utility of these methods. We believe that the proposed research will make a significant contribution to genetic epidemiology.
Showing the most recent 10 out of 160 publications
