This proposal describes a five-year plan developed for a Cancer Prevention, Control, Behavioral and Population Sciences Career Development Award. It outlines a program that integrates education, teaching and research to develop my skills in statistical genetics, molecular biology, cancer biology, human genetics and genetic epidemiology. My research plan consists of two projects: (l) higher level genetic modeling of early onset lung cancer cases and controls with admixture versus ethnicity as a covariate and (2) application of an admixture, population substructure and disequilibria testing procedure to early onset lung cancer cases, population-based controls and nuclear families. Project l uses data from a funded study on early onset lung cancer of African-Americans and Caucasians. Nine candidate loci believed to be involved in lung cancer risk along with 35 population specific markers (PSAs) for Africans and Europeans combined, are being typed for each case and a matched population and familial control. An admixture algorithm will be written, programmed and tested on cases, controls and parents to estimate individual and population admixture, using the PSAs. Logistic regression models and decision tree models (classification and regression tree) will be compared when modeling the genotypes of the candidate loci and other collected environmental/descriptive variables, in order to assess the difference between using an ethnicity variable versus an individual admixture estimate variable. Project 2 will build on the already developed admixture estimation algorithm (Project l). I will develop a complete procedure that will estimate individual and population admixture and within population substructure and will test for induced linkage disequilibrium (LD) and Hardy-Weinberg disequilibrium (HWD), for all cases, controls and nuclear families. This procedure will be tested in a computer-based simulation, using the early onset lung cancer data as a model, in order to estimate the statistical power and test characteristics such as false-positive and false-negative rates. Linkage disequilibrium mapping will also be performed on the study data. Both of these projects will address epidemiological study design issues about ethnicity and using population-based versus familial based controls. From the studies proposed, a recommendation could be made on how to utilize individual admixture information in the choice of controls for epidemiological studies. These projects will give me experience in developing methodology in statistical genetics and genetic epidemiology, while also helping me to better understand etiology of disease.
Showing the most recent 10 out of 13 publications
