全基因组选择技术改良玉米自交系出籽率的初步研究

Kernel rate is a key factor affecting the grain yield of maize. The kernel rate of maize is controlled by multiple genes with little effect, which restricts genetic manipulation in the breeding. Genome wide selection(GWS) is the newest molecular markers assisted selection(MAS) method. It can accelerate the selection breeding progress and improve the selection efficiency. We will try to use the exotic germplasm LK1 to improve the kernel rate of Qi319 by GWS. The （LK1×齐319）F2 population is the training population. The BLUP model will be made by the SNP markers analysis results of F2 single plant and the phenotypic value of the testcrosses. Single plant selection will be based on the breeding value predicted by BLUP model in the next breeding cycles. Meanwhile, the phenotypic value of the testcrosses of every breeding cycles will be recorded. Then the selection response of GWS will be calculated by the prediction value and phenotypic value. So the improvement effects of GWS will be obtained. We will also carry out the traditional phenotypic backcrossing to improve the kernel rate of Qi319, LK1 as the donor. Then we will compare the effects of GWS and phenotypic backcrossing.

出籽率是影响玉米产量的一个关键性因素。玉米出籽率受微效多基因控制、限制了育种中对其进行遗传操作的能力。全基因组选择技术(GWS)是最新发展起来的分子标记辅助选择技术，能够加快选择进度，提高选择效率。本研究尝试应用GWS技术将外源长粒种质LK1的与出籽率相关的优异基因转入到齐319中。利用（LK1×齐319）F2群体作为参考群体，根据F2单株的SNP标记分析结果和对应的测交种的表型值调查结果建立BLUP模型。以后每轮育种后代只进行SNP标记分析，然后利用BLUP模型预测的育种值选择单株。同时调查每轮育种群体测交种的表型值，计算模型对出籽率等相关性状的选择响应，从而得出GWS技术用于自交系改良的效果。同时，将LK1作为供体自交系利用传统的回交育种技术提高齐319的出籽率。比较GWS与传统回交育种的选择效果。

全基因组选择技术是最新发展起来，不需要进行QTL定位的分子标记辅助选择技术，适于低遗传力性状，能够加快选择进度，提高选择效率。本研究尝试应用全基因组选择技术将外源长粒种质LK1中与出籽率相关的优异基因转入到齐319中。利用（LK1*齐319）F2群体作为参考群体，根据F2单株的基因分型结果与对应的测交种表型值建立BLUP模型，以后的每轮育种后代只进行SNP标记分型，利用BLUP模型预测育种值并进行单株选择。并计算出模型对出籽率等相关性状的选择响应。研究结果表明，出籽率的预测响应为33.7%，高于产量、穗粒数、收获期含水量。百粒重与雄穗分枝数的预测响应高于60%。性状遗传力越高，预测效果越好；10000到55000bp的SNP密度下，性状的选择效果差异不大；参考群体在150，130，100株时，性状的预测效果差异不大。通过该项目的研究打开了国内玉米全基因组选择育种技术的先河，为玉米全基因组选择育种技术的应用奠定了基础。