Factors influencing accuracy of genomic breeding value of dairy cattle population in Thailand

Abstract

Fertility traits in dairy cows are economically important traits because these traits, directly and indirectly, affect production costs and profits. The important fertility traits of dairy cattle include the number of services per conception (NSPC), days open (DO), and calving interval (CI). However, the fertility trait is a trait that has a low heritability (h2). Selection for fertility traits is usually based on breeding value (BV) with important genetic parameters such as heritability (h2), genetic correlation (rg), phenotypic correlation (rp), etc. Genomic selection has become a new tool for genetic improvement in dairy cattle. Genomic selection can increase the rates of genetic gain through increased accuracy of estimated breeding values, reduction of generation intervals, and reduction in breeding costs. Genomic selection refers to selection decisions based on genomic breeding values (GEBV). The accuracy of GEBV depends on the number of the reference population, the heritability of the trait, the number of single nucleotide polymorphisms (SNPs), and the level of linkage disequilibrium (LD). The objectives of this study were to estimate the genetic parameter of economically important traits in dairy cattle were estimate on the dairy cattle population in Upper Northern Thailand from 1995 to 2018. The Restricted Maximum Likelihood (REML) was used to estimate the variance component of genetic parameters. And, simulate the dairy population under different level factors for genome evaluation, and analyze the impact of some factors on the accuracy of genomic breeding value of dairy cattle population in Thailand. The simulation used information on the dairy cattle population, from 1964 to 2020. Accuracy analysis used a correlation between GEBV and TBV. The 2000 and 3000 reference population sizes, heritability (h2) between 0.05 to 0.50, linkage disequilibrium (LD) between 0.05 to 0.45, and number of SNPs 20K and 40K were used. The results showed farm, month of calving, year of calving, and age of first calving were highly significant affected whole fertility traits (p<0.01). The number of services per conception was highly significantly affected to DO and CI traits (p<0.01). The heritability of NSC, DO and CI were 0.03±0.01, 0.15±0.02, and 0.16±0.02, respectively. In addition, genotypic and phenotypic correlations on whole fertility traits were positive. And the accuracy estimated by genomic data of CS tended to be better than VS. While, whereas the different numbers of SNPs tended to be similar in the accuracy of GEBV. In addition, the accuracy of low h2 and low LD is not stable.