Preliminary Study On Genome Selection By Using Haplotype In Chinese Simmental

Genomic selection(GS) is the marker assisted selection on a genome wide scale, which combines phenotypic and markers’ information to calculate genomic estimated breeding values(GEBVs). GS can shorten the generation interval by selecting individuals at the early stage of life, accelerate genetic progress and decrease the breeding cost. Therefore, genomic selection has been become a popular research hotspots for the animal breeders. Before conducting genomic selection using SNP markers on whole chromosome, we assumed the SNP and QTL were in linkage disequilibrium, and whole SNPs were independence. We accumulated the SNP indicator variable multiplying marker effects to estimate the genomic breeding value. However, in fact, SNPs were in linkage disequilibrium extensively in cattle genome. Therefore, we used haplotype in genomic selection to estimate the genomic breeding value.Our study used 1141 Chinese Simmental cattle that were originated from Wulagai, Inner Mongolia, which were genotyped with high density 770 K Beadchip. We used the 501951 SNPs that passed quality control to construct haplotype block. Haplotypes were constructed using three threholds including D’≥0.45,D’≥0.55,D’≥0.65 and D’≥0.75 to explore the optimal threshold in Simmental population. Through constructing the haploblock matrix by these haplotype block information,we used GBLUP, BayesA, and BayesCπ to estimate genomic breeding value in the trait of live weight, carcass weight, dressing percentage and meat percentage. We compared predictive accuracy of GEBVs estimated by both SNPs and haploblock. In our study, we obtained the highest the predictive accuracyusing GBLUP, BayesA and BayesCπ, when the threhold was D’≥0.55. For meat percentage trait, GBLUP had the highest predictive realized accuracy for GEBVs. While, BayesA had the highest predictive accuracy, when the threhold was D’≥0.55. However, Bayescπ had the highest predictive accuracy when D’≥0.65.Genomic selection by haplotype method can improve both in the computational time and increase the accuracy of estimation of breeding value. There is a high value of application in animal breeding.