Statistical Methods for Family-Based Association Studies

描述（由申请人提供）：越来越多的证据表明，全基因组关联研究（GWAS）是鉴定与常见人类疾病有关的基因的有力方法。GWAS可以使用基于人口的设计或传统的基于家庭的设计。基于家庭的GWAS的最大优点之一是它对可能增加假阳性率的人口分层的影响具有稳健性。然而，这种健壮性也可能导致功率的损失。因此，在基于家庭的设计下，GWAS需要更强大的关联检验，对群体分层具有稳健性。另一方面，模拟研究以及对几种常见疾病的遗传结构的研究表明，因果变异包括常见和罕见。新技术允许对大量个体的部分基因组进行测序，或者在未来，对整个基因组进行测序。用于检测常见变异关联的统计方法在检测罕见变异关联时可能不是最佳的。因此，迫切需要开发强大的统计方法来检测基于家族的序列数据中的罕见变异。本项目将探索新的统计方法和可行的算法来绘制基于家族的GWAS的复杂疾病基因图谱。该项目的第一个具体目标是为基于家庭的GWAS开发一种更强大的单标记两阶段联合分析，该分析对人口分层具有强大的作用。第二个具体目标是开发一种新的关联测试，可以检测基于家族设计的罕见变异。第三个具体目标是使用广泛的模拟研究来比较所提出的方法与现有方法，并将所提出的方法应用于选定的基于家庭的GWAS数据集。该项目的最后一个具体目标是为新开发的方法开发计算机软件，并免费向科学界发布软件。这些新的统计方法和用户友好工具的发展将有助于研究人员对复杂遗传性状的基因进行基因组定位。新的、可靠的统计方法将极大地造福科学界。

DESCRIPTION (provided by applicant): There is increasing evidence that genome-wide association studies (GWAS) represent a powerful approach in the identification of genes involved in common human diseases. GWAS may use either population-based designs or traditional family-based designs. One of the biggest advantages of family-based GWAS is its robustness to possible effects of population stratification, which can inflate the false positive rate. However, this robustness can also lead to a loss of power. Thus, more powerful association tests that are robust to population stratification are needed for GWAS under family-based designs. On the other hand, simulation studies as well as studies of the genetic architectures of several common diseases suggest that causal variants include both common and rare. New technologies allow for sequencing of parts of the genome-or, in the future, the whole genome-of large groups of individuals. Statistical methods developed to detect associations of common variants may not be optimal in detecting associations of rare variants. So there is a great need to develop powerful statistical methods to detect rare variants for family-based sequence data. This proposed project will explore novel statistical methods and feasible algorithms to map complex disease genes for family-based GWAS. The first specific aim of this project is to develop a more powerful single-marker two-stage joint analysis for family-based GWAS that is robust to population stratification. The second specific aim is to develop a new association test that can detect rare variants under family-based designs. Using extensive simulation studies to compare the proposed methods with existing methods and applying the proposed methods to selected family-based GWAS data sets are the third specific aim. The last specific aim of this project is to develop computer software for the newly developed methods and release the software to the scientific community at no charge. The developments of these novel statistical methods and the user friendly tools generated from this project will aid researchers in genomic localization of genes that contribute to complex genetic traits. New, sound statistical methods will greatly benefit the scientific community.