RISK ESTIMATION AND MODIFIER GENES IN INHERITED CANCER

点击翻译按钮获取中文摘要

DESCRIPTION (Adapted from investigator's abstract): During the past decade, progress has been made in localizing and identifying specific genes such as BRCA1 and p16 which when altered can confer a markedly increased susceptibility to relatively common cancers. Although variations at these genes apparently increase risk of cancer at a variety of anatomical sites, there is controversy about the magnitude of these risks. Risks derived from population-based studies have produced lower estimates of risk than those estimated from studies of single large families or series of multiple high-risk families. Whether these differences reflect inherent biases in the methods used for risk estimation or represent the effects of modifying loci segregating in high-risk families is not obvious. Statistical methods for estimation of gene effects are relatively underdeveloped. Additionally, there have been no studies that examine the feasibility of mapping hypothesized modifier loci in the context of a total genome search. This project seeks to examine the efficiency and bias associated with various sampling designs and statistical methods for estimation of cancer risks due to such genes. Monte-Carlo simulation and existing datasets encompassing the three approaches described above will be used for the evaluation of a variety of methods. Several designs for a) testing the effect on risk modification of specific candidate loci; b) detecting the presence of residual genetic effects; and c) examining the power of different sampling designs for genome-wide screens to localize modifier loci also will be examined.