Construction And Updating Strategies Of Pig Genome Selection Population

In October 2017,the Ministry of Agriculture and Rural Affairs of the People’s Republic of Chinalaunched the Pig Genome Selection Project in Chongqing,which bringing China’s pig breeding work into line with international standards.Currently,all participating enterprises have been establishing the reference groups.However,the application of genomic selection technology is a systematic project,involving the construction and update of reference group,the composition of candidate group and compatibility of different genome chips in joint genetic assessment.At present,for these problems,our country has not been relatively perfect technical program.Based on this,this study will focus on the establishment of a series of technical schemes in the implementation process of pig genome selection.The main researches are as follows:1.Research on reference group construction strategy and update schemeThis study simulated the population structure of 1200 purebred pigs in China for 20 generations.Taking the 20 th generation as the candidate group and the 5th to 19 th generation as the reference group,the effects of different reference group sizes,relationship between reference group individuals and between reference group and candidate group on the accuracy of GEBV were studied.The results showed that the accuracy was positively correlated with the size of the reference group within the same generation(r=0.9366,p<0.05).There was an extremely significant negative correlation with the number of generation intervals between the reference group and the candidate group(r=-0.9267,p<0.01).In addition,the accuracy first increased and then decreased with the accumulation of the reference group between generations.The first three generations increased obviously,and then the growth rate weakened.When the reference group accumulated more than 7 generations,the accuracy began to decline.Therefore,this study suggests that when constructing the reference group for the first time,individuals with the closest relatives to the candidate group should be selected,and the reference group should contain all the genetic variations of the population as far as possible.In addition,the accumulation of the reference group between generations is maintained within 3 to 7 generations,that is,we should get rid of the old individuals when the news are included into the reference group,and this work can be started from the 4th generation at the earliest and the 8th generation at the latest.2.Study on candidate group determination scale and gender ratioThis study simulated the population structure of 1200 purebred pigs in China for 5 generations,and studied the effects of different heritability(0.1,0.3,0.5,0.8),different determination ratios(20%,30%,50%)and different sex ratios on the accuracy of GEBV and genetic gain.The results showed that,if the proportion of genotyping individuals accounted for 20% of the candidate group,the traits with different heritability could be genotyped according to the sex ratio of male and female 1:1.If the proportion was set to 30%,and the traits are of low heritability(0.1),the sex ratio of male to female is according to 1:1.If the traits are medium or high heritability,the ratio of male to female is according to 1:1,1:2 and 2:1which maybe achieve more higher genetic progress.If the genotyping proportion is upto 50%,for low heritability traits(0.1),the proportion of sows from the all genotyped individuals should not be less than25%,and for the medium and high heritability traits,the optimal choice for the male-female ratio is 1:1,which maybe obtain the best genetic progress.3.Influence of basic data of different years on the parameter estimation and the accuracy of genome estimation breeding valueIn this study,the effects of 7,5,3 and 1 years basic data on the genetic parameters and GS of 6growth and carcass traits of Whiteshire using the phenotype data from 2011 to 2017 and the genotyping data of 1633 pigs from a large pig farm in China.The results showed that the genetic variance of each trait was too high and estimated with only one year’s data and the reference value was low.The basic data of other years had no significant effect on variance components,genetic correlation and heritability of traits,and had no significant effect on the accuracy of GEBV.4.Applied research on gene chip imputationThis study chooses two of the most widely used chips on the market in our country Gene Seek50 k and "ZXYH",imputes each other by Fimpute,and compares the influence of six reference group design strategies on filling accuracy when imputating from "ZXYH" to Gene Seek50 k.The results show that the filling accuracy range of 0.962 ～ 0.985,of which the reference group from a single group with pedigree information has the highest filling accuracy of 0.985.The application effect of the original dataset(2000head),the filled dataset filling from the original dataset(2000 head),and the mixed chip from original data set(1000 head)and the filled data set(1000 head)selected randomly was studied.The results showed that there was no significant difference in the accuracy and unbiasedness of GEBV from the three data sets.And the Spearman correlation of GEBV of the three dataset is more then 0.99.It is proved that the chips with different markers have good compatibility after imputating and can be used for the joint genetic assessment of Chinese breeding pigs.