论文部分内容阅读
百粒重是大豆产量的重要构成因子,在一定条件下与产量呈显著正相关。百粒重是一个复杂的数量性状,用传统的育种方法其遗传增益不明显。本研究对280份大豆品种多年多点田间鉴定,通过混合线性模型预测获得品种百粒重的最佳线性无偏预测值。同时利用分布在大豆全基因组的5361个SNP标记鉴定参试品种基因型,结合随机回归最佳线性无偏预测模型和交互验证方法,探讨了群体构成方式对大豆百粒重的全基因组选择预测准确度的影响。结果表明,大豆百粒重的全基因组选择预测准确度变化范围为–0.15~+0.75;群体构成方式对百粒重的预测准确度影响明显;亚群内的预测准确度(0.24~0.75)高于亚群间(-0.15~+0.29);当群体间遗传距离由0.1566增加到0.2201时,预测准确度下降27.87%;相比随机构建的训练群体,基于群体遗传结构构建的训练群体能将百粒重的预测准确度提高2.34%。本研究明确了大豆百粒重的全基因组选择预测准确度,阐明了群体结构对大豆百粒重的全基因组选择预测准确度的影响,为大豆分子育种提供了新的思路和方法。
100-grain weight is an important component of soybean yield, under certain conditions and yield was significantly positively correlated. One hundred kernel weight is a complex quantitative trait, whose genetic gain is not obvious with traditional breeding methods. In this study, 280 soybean cultivars were identified for many years in field and the best linear unbiased predictive value was obtained by hybrid linear model prediction. At the same time, 5361 SNPs in soybean genome were used to identify the genotypes of the tested cultivars. Combined with the best linear unbiased prediction model and the cross-validation method, the whole genome selection of soybean population was analyzed. Degree of influence. The results showed that the prediction accuracy of whole-genome selection of 100-grain weight ranged from -0.15 to + 0.75. The population composition method had a significant effect on the prediction accuracy of 100-grain weight. The prediction accuracy within the subpopulation ranged from 0.24 to 0.75 (-0.15 ~ +0.29). When the genetic distance between populations increased from 0.1566 to 0.2201, the accuracy of prediction decreased by 27.87%. Compared with the randomly-constructed training groups, the training groups based on the population genetic structure could produce 100 Prediction accuracy of grain weight increased by 2.34%. In this study, the accuracy of genome-wide selection prediction of soybean 100-kernel weight was clarified, and the influence of population structure on the prediction accuracy of whole-genome selection of soybean 100-kernel weight was clarified. This provided a new idea and method for soybean molecular breeding.