基于gBLUP方法及Cross-validation大豆表型精准预测研究  

ReseachAccurcay Prediction of Soybean Phenotype by gBLUPMethod and Cross-validation

作　　者：唐友[1,2] 郑萍[2] 张继成[2] 

机构地区：[1]黑龙江财经学院,哈尔滨150076 [2]东北农业大学,哈尔滨150081

出　　处：《青岛大学学报（自然科学版）》2017年第1期56-59,共4页Journal of Qingdao University(Natural Science Edition)

摘　　要：为了实现提高产量和抵抗病害等能力的目的,需要提高育种水平,通过设计交差验证(Cross-Validation)实验进行大豆基因型和表型数据的分组处理,根据数据的个体和mark的数量进行合理分配,采用gBLUP(genomic Best Linear Unbiased Prediction)方法进行表型预测。根据对大豆数据多个性状通过不同分组的对比来得到精确值的范围,为后续的育种分析提供依据。对于只有大豆基因型数据而没有表型数据的情况,需要模拟表型,根据设定遗传力和模拟位点的个数(NQTN)进行模拟,然后再进行不同分组获取精准值,这样扩大了大豆数据的预测灵活性。The breeding level should be improved for achieve the purpose of improving yield and resistance to disease.GBLUP（genomic Best Linear Prediction Unbiased）method was used to predict the phenotype,through the design of cross validation（Cross-Validation）experimental grouping of soybean genotypes and phenotypic data,the reasonable distribution of data according to the number of individuals and mark.The range of the precise values was obtainedthrough the comparison of different groups based on the comparison of the different characters of soybean data to provide the basis for subsequent breeding analysis.The phenotype was simulatedaccording to heritability and the number of loci（NQTN）,and then get the precise value of different groups,so as to expand the flexibility of soybean data forecast.

关 键 词：交叉验证 表型预测 gBLUP 遗传力 

分 类 号：Q3-3[生物学—遗传学]