考虑基因互作时的复杂性状遗传机制解析、预后方法与应用研究

Identifying the genetic mechanisms of complex traits, and then guiding genetic improvement and animal production are the current main tasks. In recent years, we have developed a variety of new methods, software and platform involving genome-wide association studies (GWAS) and genomic selection (GS). We also used these tools to reveal the genetic basis and perform genetic improvement of milk yield. However, we found that ignoring the genetic interaction, would significantly affect the accuracy and reliability of GWAS and GS results. Meanwhile, it is also very important to evaluate the animals themselves lifetime performance (additive effects + interaction effects) of complex traits based on genomic prognosis analysis (GP). Motivated by these concerns, the project intends to perform following further work based on our preliminary methods. Firstly, taking full account of the presence of genetic interaction in GWAS, GS and GP, we will develop a series of new methods with low false positive and false negative rates, high prediction and prognosis accuracy, and fast computing speed, by constructing the most appropriate model and optimizing the calculation methods. Simulation and validation study will also be performed based on cattle, pigs and other real datasets to examine the effectiveness and feasibility of these methods. Secondly, we will carry out complementary genetic variation detection and phenotype measurement, and then apply the new methods to interpret the molecular genetic architecture of milk production, especially for the functional productive life and its related traits, and conduct GS and GP. Implementation of this project will play a key role in understanding the molecular genetic basis and genomic prognosis of animal complex traits.

揭示复杂性状的遗传机制并据以指导改良、生产是当前的主要任务。近年来，我们围绕基因组关联分析（GWAS）和基因组选择（GS），开发了多种新的方法、软件、平台，并用到了奶牛产奶量的遗传机制解释与遗传改良中。但研究中我们发现无视基因互作，将大大影响GWAS和GS结果的准确性和可靠性。同时，对动物本身复杂性状的终生表现（加性效应+互作效应）进行预后分析（GP）也有重要意义。为此，我们拟进一步开展如下工作：①在GWAS、GS、GP中充分考虑基因互作的存在，通过最宜模型的构建、算法的优化，开发出假阳性率与假阴性率低、估计与预后准确度高、运算速度快的系列新方法，进而通过模拟研究与牛、猪等实际数据检测其有效性与可行性。②补充进行基因组变异检测和性状表型测定，进而应用新方法揭示产奶量尤其是母牛使用寿命及其相关性状的遗传机制，进行GS与GP。本项目的实施对于动物复杂性状的遗传机制解释、预后研究具有重要的价值。

在全基因组关联分析、基因组预后研究中如果无视基因互作，将大大影响结果的准确性和可靠性。为此，我们开展了如下工作：①在GWAS、GP中充分考虑基因互作的存在，通过最宜模型的构建、算法的优化，开发了假阳性率与假阴性率低、估计与预后准确度高、运算速度快的系列新方法PPGWAS、GIGP等，进而通过模拟研究与牛、猪等实际数据检测证明了方法有效性与可行性。②共完成了2092头奶牛参考群的基因组变异检测和性状表型测定，进而应用PPGWAS等新方法初步揭示母牛使用寿命及其相关性状的遗传机制；对于乳蛋白量、乳脂量和体细胞评分等性状，GIGP法的预测准确度显著高于BayesB和GBLUP法，特别是针对反映乳房炎疾病的体细胞评分，GIGP法比GBLUP法和BayesB分别高9%和10%。本项目的研究成果对于动物复杂性状的遗传机制解释、基因组选择研究具有重要的价值。同时，对动物本身复杂性状的终生表现（加性效应+互作效应）进行预后分析也有重要意义。