| With the innovation of breeding technology of livestock and poultry,breeders have higher and higher requirements on the accuracy of genetic assessment.Due to the late start of modern livestock and poultry breeding in China and the lack of breeding experience in the early stage,there are a large number of genealogical errors in actual production,which affect the accuracy of traditional BLUP model in estimating animal breeding values and greatly reduce the genetic progress in one breeding cycle.Genomic selection has been widely used in the genetic assessment of livestock and poultry.It USES genome-wide high density SNP markers to estimate the breeding value of individual genome.The methods of genome selection are divided into "multi-step method" and "one-step method".Because the one-step method USES a large amount of genealogical information during its implementation,genealogical errors will affect the accuracy of the one-step method in predicting genomic breeding values.With the increase of genealogical errors,it is seldom reported whether the one-step method still has an advantage over the multi-step method.The purpose of this study was to investigate the influence of different degrees of genealogical errors on the prediction of genomic breeding values by one-step and multi-step methods.In this study,simulated population and actual population were used as research objects to explore the influence of different degrees of genealogical errors on the accuracy of genome selection.Heritability in different traits,and the reference group of number and number of QTL,use traditional BLUP method,the "steps" and "one-step" respectively to investigate the influence of different degree of pedigree error for genome selection accuracy,it is concluded that the number of heritability and QTL and to different degrees of different error predicting breeding value of genome selection accuracy change rule has no obvious influence.When the reference group was 3000 and the genealogy error rate reached about 24%,the reliability of the one-step method to predict the genomic breeding value was lower than that of the multistep method to predict the breeding value.When the reference group was 5000 and the genealogical error rate reached about 8%,the reliability of one-step method to predict the genomic breeding value would be lower than that of multi-step method to predict the genomic breeding value.This study used different methods to detect and correct pedigree errors,and evaluated the ability of BLUPf90 software and Prest software to detect and correct pedigree errors.In the simulated population,it is completely feasible to correct genealogy through genome data,in which both can detect more than 99% genealogical errors.Prest software is better than BLUPf90 software in correcting genealogical errors.However,when the number of markers and the number of detected individuals are large,the running time of Prest software will increase exponentially. |
