【作者】 卫巍；

【导师】 童春发；

【作者基本信息】 南京林业大学 ， 林木基因组与生物信息学， 2009， 硕士

【摘要】 QTL功能作图统计方法将动态性状的发育机制组装到QTL作图统计模型中,目前已经建立了近交群体(回交群体、F2群体等)的统计模型。林木属于异交群体,有许多自身的生物学特性,不能简单套用近交群体的作图方法,必须建立适合林木自身特性的QTL作图方法。本文针对林木二倍体全同胞家系,考虑双亲的QTL基因型一个纯合,另一个杂合的情况,将全同胞家系所有的标记分离类型对和连锁相类型考虑到模型中,并将Logistic生长曲线组装到模型中,建立了林木F1代群体的QTL功能作图统计方法,使用EM算法获得模型未知参数向量的极大似然估计。从模拟结果来看,即使在个体性状值向量的维数、遗传力和群体大小都较小的情况下,无论是QTL位置还是各个参数估计的均值都与真值非常接近,且误差均方根都比较小,这说明模型在定位QTL和参数估计方面有明显的效力;当个体性状值向量的维数、遗传力水平和群体大小任何一个因素提高时都能明显改善参数估计的精确度。通过分析一组杨树的实例数据,使用林木QTL功能作图的方法对影响杨树单株生根总数和最大根长两个性状的QTL进行定位来寻找与生根能力相关的QTL,总共搜索到三个显著的控制不定根发育轨迹的QTL。我们编写了林木全同胞家系QTL功能作图软件——FsQtlFunMap 1.0,该软件在计算速度上有明显的优势,可进行相关的统计分析和图形绘制。更多还原

【Abstract】 Functional mapping of quantitative trait loci (QTL) has embedded the biological mechanisms and processes of dynamic traits in the statistical method of QTL mapping. Although the functional mapping statistical model for mapping QTL has been well developed in inbred lines, it has not been proposed in outbred species, e.g. a full-sib family, in which the number of alleles may vary over loci and the linkage phases are usually unknown between loci. Since forest trees have several unique biological characteristics and most of them are outbred species, we should not directly apply the functional mapping model for inbred species to them.The objective of this study is to develop a functional mapping statistical model for a full-sib family in forest trees. We suppose that the QTL genotype of one parent is heterozygous but the other is homozygous, and consider all pairwise combinations of marker segregation types and all possible linkage phases between mark loci in a diploid full-sib family into our model. A maximum likelihood approach based on a logistic-mixture model, implemented with EM algorithm, was developed to provide the estimates of QTL positions, QTL effects, and other model parameters responsible for growth trajectories.According to the results of a large number of simulations, we can infer that the estimation precision of the QTL effects and position is dependent on dimensions of measured variable, sample size and heritability. Even though when the dimension of measured variable, heritability and sample size are all smaller, the mean of the estimation of QTL location and each effective parameters approximates to the true value very closely and all the SMSEs of MLEs are also smaller, increased one of the factors (i.e. dimensions of measured variable, sample size and heritability) can decrease the SMSEs of MLEs of partial or all parameters obviously. All that indicate our model was robust for identifying the QTL position and estimating the parameters of QTL effects and residual (co)variance.We use an example of poplar to demonstrate the validation of our functional mapping method for mapping QTL affecting growth trajectories in practice. The high-density genetic linkage map has been constructed with a mix of marker systems. Functional mapping identified two QTL affecting MRL (maximal root length) on the linkage 3 and one QTL affecting TNR (total root number per cutting) on the linkage 8.We have developed Windows software, FsQtlFunMap 1.0, which was written in Visual C++ 6.0 to implement our functional mapping method for a full-sib family. FsQtlFunMap 1.0, which was used to map QTL in a full-sib family in forest trees, can do QTL statistical analysis and draw relevant figures. It has the remarkable advantage of speed for calculating.更多还原