A general and accurate approach for computing the statistical power of the transmission disequilibrium test for complex disease genes
A general and accurate approach for computing the statistical power of the transmission disequilibrium test for complex disease genes
复制标题
DOI：
10.1002/gepi.1018
复制
发表时间：
2001-07-01
影响因子：
2.1
通讯作者：
Deng, HW
中科院分区：
文献类型：
--
作者：
Chen, WM;Deng, HW
传递不平衡检验（Transmission disequilibrium test，TDT）是一种以核心家系为基础的分析方法，可以在存在关联的情况下检验连锁。它在理论研究和实际应用中得到了广泛的关注，在这两种情况下，TDT的功率计算的准确性和通用性是至关重要的。尽管进行了广泛的调查，以前的方法来计算统计功率的TDT既不准确，也不一般。在本文中，我们开发了一个通用的和高精度的方法来分析计算的TDT的功率。我们比较我们的方法与其他几个最近的论文，所有对计算机模拟得到的结果的结果。我们表明，从我们的方法计算的结果是更准确的比或至少相同的那些从其他方法。更重要的是，我们的方法可以处理各种情况，包括（1）由一个或多个子女组成的家庭，并有任何配置的受影响和非受影响的同胞兄弟姐妹，（2）通过父母的情感状态确定的家庭，（3）任何混合样本与不同类型的家庭（1）和（2），（4）标记基因座不是一个疾病易感基因座，（5）标记基因座是一个疾病易感基因座。（5）存在等位基因异质性。我们实现了这种方法在一个用户友好的计算机程序：TDT功率计算器。它的应用程序进行了演示。这里开发的方法和程序应该是有意义的理论家，以准确地调查在各种情况下的TDT的统计能力，并为经验遗传学家计划使用TDT的有效研究。(C)2001 Wiley-Liss，Inc.
Transmission disequilibrium test (TDT) is a nuclear family-based analysis that can test linkage in the presence of association. It has gained extensive attention in theoretical investigation and in practical application; in both cases, the accuracy and generality of the power computation of the TDT are crucial. Despite extensive investigations, previous approaches for computing the statistical power of the TDT are neither accurate nor general. In this paper, we develop a general and highly accurate approach to analytically compute the power of the TDT. We compare the results from our approach with those from several other recent papers, all against the results obtained from computer simulations. We show that the results computed from our approach are more accurate than or at least the same as those from other approaches. More importantly, our approach can handle various situations, which include (1) families that consist of one or more children and that have any configuration of affected and nonaffected sibs; (2) families ascertained through the affection status of parent(s); (3) any mixed sample with different types of families in (1) and (2); (4) the marker locus is not a disease susceptibility locus; and (5) existence of allelic heterogeneity. We implement this approach in a user-friendly computer program: TDT Power Calculator. Its applications are demonstrated. The approach and the program developed here should be significant for theoreticians to accurately investigate the statistical power of the TDT in various situations, and for empirical geneticists to plan efficient studies using the TDT. (C) 2001 Wiley-Liss, Inc.