报告题目: Bivariate Genomic Selection for Boosting Predictability of Low-Heritability Traits

报告人：贾震宇 助理教授

报告时间：2018年7月13日（周五）14:30

报告地点：逸夫楼C座314会议室

摘要：

Due to the complexity of modeling, the multivariate genomic
selection (GS) models have not been adequately studied and their potential advantages over univariate GS models remain unclear. In this study, we developed a novel bivariate (2D) BLUP-HAT GS method by leveraging computationally efficient HAT method. The advantages of using the new bivariate GS model over the univariate (1D) BLUP-HAT GS model have been demonstrated by analyzing 4 conventional traits and 1000 metabolomic traits of rice. The results indicated that (1) the 2D GS analysis generally produces higher predictability than the 1D GS analysis, and (2) lowheritability traits may significantly benefit from the 2D GS analysis when paired with a high-heritability trait.

报告人简介：

贾震宇博士于1998年和2006年分别在武汉大学和加州大学河滨分校获学士和博士学位。后在加州大学尔湾分校病理系、阿克伦大学统计系、俄亥俄西北医科大学医学系和加州大学河滨分校植物科学系工作，长期从事各类遗传学数据的深度分析和生物统计与生物信息学教学和科学研究。担任Journal of Data Mining in Genomics & Proteomics、 Frontiers Plant Science等杂志的编委或编辑。主持科研项目3项。在Cancer Research、 Genome Biology、Bioinformatics 和Genetics 等刊物发表SCI论文50余篇。目前主要研究全基因组选择中的统计遗传学问题。