STATISTICAL METHODS FOR GENETIC EPIDEMIOLOGY

描述：这项新提案的目标是继续 开发和实施新的遗传基因统计方法 癌症、变态反应等复杂慢性病的流行病学研究 和精神障碍。重点将放在开发新的 多变量生存数据的统计分析方法 通常在收集家庭数据以解决以下问题时 检测家族聚集性并识别不同的遗传亚型； 用多点连锁方法定位复杂疾病的致病基因 或关联研究方法；将遗传异质性纳入 遗传连锁数据的分析。对于这些具体目标中的每一个，我们 将对现有的统计方法进行评估并指出其局限性 以及这些因素对得出有效推论的影响；开发和研究两者 在分析和经验上，适当的新统计方法 要解决这些问题，请将这些新的统计方法应用于实际数据 各种复杂疾病的集合，促使这一提议 研究、开发和分发软件以实施这些新的统计数据 技巧。这里提出的工作将像我们过去所做的那样做出贡献， 一般遗传流行病学领域的统计方法和 提供对各个临床领域的洞察 用于说明这些新方法的数据集。

DESCRIPTION: The objective of this renewed proposal is to continue developing and implementing new statistical methodology for genetic epidemiologic studies of complex chronic diseases including cancer, allergy and psychiatric disorders. The focus will be on the development of new statistical methods for analyzing multivariate survival data which occurs frequently when family data are collected to address issues such as detecting familial aggregation and identifying distinct genetic subtypes; mapping disease genes for complex diseases using multipoint linkage method or the association study method; incorporating genetic heterogeneity into the analysis of genetic linkage data. For each of these specific aims, we will evaluate existing statistical methods and point out their limitations and the impact of these on drawing valid inferences; develop and study both analytically and empirically, new statistical methods that are appropriate to address these issues; apply these new statistical methods to actual data sets of a variety of complex diseases which have motivated this proposed research; develop and distribute software to implement these new statistical techniques. The work proposed here will contribute, as we have in the past, statistical methodology to the field of genetic epidemiology in general and offer insight into each of the clinical areas represented by the various data sets used to illustrate these new methods.