Epidemiology of Putative Causal Variants in the Multiethnic Cohort

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): In response to RFA-HG-07-014, we propose to use the Multiethnic Cohort (MEC) study to characterize the "epidemiologic architecture" of putative causal variants identified in large-scale genomic association studies for a wide range of complex traits (chronic diseases, intermediate phenotypes and behavioral risk factors) across racial/ethnic populations. We have established a large biorepository of blood and urine (N=67,000) and cryopreserved lymphocytes (N=15,000) linked to extensive, prospectively collected risk factor (e.g., diet, smoking, physical activity), biomarker and clinical data, for five racial/ethnic groups in the MEC. This cohort study of over 215,000 men and women in Hawaii and California is unique in that it is population-based and includes large representations of older adults (45-75 yrs at baseline) for five US racial/ethnic groups (Japanese Americans, African Americans, European Americans, Latinos and Native Hawaiians) at varying risks of chronic diseases. We propose to study: 1) diseases for which we have DNA available for large numbers of cases and controls (breast, prostate, and colorectal cancer, diabetes, and obesity); 2) important cancers that are less common (e.g., lung, pancreas, endometrial cancers, NHL) but for which we propose to pool our data with other funded groups; 3) common traits that are risk factors for these diseases (e.g., body mass index/weight, waist-to-hip ratio, height) and 4) relevant disease-associated biomarkers (e.g., fasting insulin and lipids, steroid hormones). Our specific aims are: 1) To determine the population-based epidemiologic profile (allele frequency, main effect, heterogeneity by disease characteristics) of putative causal variants in the five racial/ethnic groups in the MEC; 2) for variants displaying effect heterogeneity across ethnic/racial groups, we will utilize differences in LD to identify a more complete spectrum of associated variants at these loci; 3) investigate gene x gene and gene x environment interactions to identify modifiers; 4) examine the associations of putative causal variants with already measured intermediate phenotypes (e.g., plasma insulin, lipids, steroid hormones); and 4) for variants that do not fall within known genes, start to investigate their relationships with gene expression and epigenetic patterns in small genomic studies. We will coordinate the selection of these variants and endpoints, analytical methods, data analyses, and rapid reporting of results with other cohorts/clinical trials funded through this RFA.