Pleiotropic QTL analysis
Pleiotropic QTL analysis
复制标题
DOI：
10.2307/2533998
复制
发表时间：
1998-03-01
期刊：
影响因子：
1.9
通讯作者：
Grimsley, N
中科院分区：
文献类型：
--
作者：
Mangin, B;Thoquet, P;Grimsley, N
利用遗传标记检测影响数量性状的基因的统计学方法已发展得很好，可用于单个性状的分析。在实践中，许多实验数据包含对多个相关性状的观察，现在需要允许联合分析所有性状的方法。将最大似然法推广到多性状分析是一种很好的方法，但由于同时估计的参数数量增加而增加了复杂性，当性状数量较多时，可能会限制其实际应用。我们提出了一种基于两个独立步骤的替代方法。第一步是估计性状的（协）方差矩阵，并使用此估计来获得与性状相关的典型变量。第二步是对每个典型变量应用单性状最大似然法，并将结果联合收割机组合。在假定的多效QTL效应的局部渐近框架中工作，即，对于效应太小而不能确定检测的多效QTL，我们证明了典型变量的组合分析渐近等价于多性状最大似然分析。还给出了多效QTL定位的阈值。检测QTL的概率并不总是通过增加更多相关性状来增加。作为一个例子，一个多性状分析与两个变量的功率和功率的单性状分析的理论比较。以番茄青枯病多基因抗性的实验数据为例，说明了典型变量的组合分析。
Statistical methods for the detection of genes influencing quantitative trait (QTLs) with the aid of genetic markers are well developed for the analysis of a single trait. In practice, many experimental data contain observations on multiple correlated traits and methods that permit joint analysis of all traits are now required. Generalisation of the maximum likelihood method to a multitrait analysis is a good approach, but the increase in complexity due to the number of parameters to be estimated simultaneously, could restrain its practical use when the number of traits is large. We propose an alternative method based on two separate steps. The first step is to estimate the (co)variance matrix of the traits and use this estimate to obtain the canonical variables associated to the traits. The second step is to apply a single-trait maximum likelihood method to each of the canonical variables and to combine the results. Working in a local asymptotic framework for the effects of the putative pleiotropic QTL, i.e., for a pleiotropic QTL whose effect is too small to be detected with certainty, we prove that the combined analysis with canonical variables is asymptotically equivalent to a multitrait maximum likehood analysis. A threshold for the mapping of the pleiotropic QTL is also given. The probability of detecting a QTL is not always increased by the addition of more correlated traits. As an example, a theoretical comparison between the power of a multitrait analysis with two variables and the power of a single-trait analysis is presented. Experimental data collected to study the polygenic resistance of tomato plants to bacterial wilt are used to illustrate the combined analysis with canonical variables.