Validation of single-step genomic predictions using the linear regression method for milk yield and heat tolerance in a Thai-Holstein population

Abstract

Background and Aim: Genomic selection improves accuracy and decreases the generation interval, increasing the selection response. This study was conducted to assess the benefits of using single-step genomic best linear unbiased prediction (ssGBLUP) for genomic evaluations of milk yield and heat tolerance in Thai-Holstein cows and to test the value of old phenotypic data to maintain the accuracy of predictions. Materials and Methods: The dataset included 104,150 milk yield records collected from 1999 to 2018 from 15,380 cows. The pedigree contained 33,799 animals born between 1944 and 2016, of which 882 were genotyped. Analyses were performed with and without genomic information using ssGBLUP and BLUP, respectively. Statistics for bias, dispersion, the ratio of accuracies, and the accuracy of estimated breeding values were calculated using the linear regression (LR) method. A partial dataset excluded the phenotypes of the last generation, and 66 bulls were identified as validation individuals. Results: Bias was considerable for BLUP (0.44) but negligible (–0.04) for ssGBLUP; dispersion was similar for both techniques (0.84 vs. 1.06 for BLUP and ssGBLUP, respectively). The ratio of accuracies was 0.33 for BLUP and 0.97 for ssGBLUP, indicating more stable predictions for ssGBLUP. The accuracy of predictions was 0.18 for BLUP and 0.36 for ssGBLUP. Excluding the first 10 years of phenotypic data (i.e., 1999-2008) decreased the accuracy to 0.09 for BLUP and 0.32 for ssGBLUP. Genomic information doubled the accuracy and increased the persistence of genomic estimated breeding values when old phenotypes were removed. Conclusion: The LR method is useful for estimating accuracies and bias in complex models. When the population size is small, old data are useful, and even a small amount of genomic information can substantially improve the accuracy. The effect of heat stress on first parity milk yield is small.

Accounting for population structure in genomic predictions of Eucalyptus globulus

“…Our results showed only a small improvement in stability between ABLUP and HBLUP in the single population , whereas other studies have demonstrated more marked increases caused by the addition of genomic information (e.g. Cesarani, Biffani, et al 2021 ; Kluska et al 2021 ; Sungkhapreecha et al 2021 ). Including MF improved our HBLUP model stability by 0.01 in the single population and 0.03 in the joint population , while it had mixed impacts on stability across 4 traits in Kluska et al (2021) .…”

“…In Bradford et al ’s (2019) simulation study, an incomplete population displayed large positive bias by ABLUP, a much smaller negative bias by HBLUP, and effectively no bias by HBLUP with MF. Cesarani, Biffani, et al (2021) and Sungkhapreecha et al (2021) both found that biases present in ABLUP models were greatly diminished using HBLUP on the same populations. In contrast, Granado-Tajada et al (2020) found that bias in dairy sheep evaluations was greater for HBLUP than ABLUP, although the difference was not statistically significant.…”

“… Cappa et al 2017 , 2018 ; Thavamanikumar et al 2020 ; Callister et al 2021 ). ACC has been shown to be higher in HBLUP than ABLUP models using the LR method in livestock populations (Cesarani, Biffani, et al 2021 ; Kluska et al 2021 ; Sungkhapreecha et al 2021 ) and by comparing true and estimated (G)EBVs in simulated experiments ( Garcia-Baccino et al 2017 ; Bradford et al 2019 ; van Grevenhof et al 2019 ). Addition of MFs to HBLUP models did not improve ACC in our study, which differs from results of simulated populations ( Garcia-Baccino et al 2017 ; Bradford et al 2019 ; van Grevenhof et al 2019 ) that showed substantial increases in ACC with the inclusion of MF.…”

“…Addition of genomic information has corrected strong over-dispersion found for ABLUP models in actual datasets (e.g. milk yield in Sungkhapreecha et al 2021 ; direct weaning weight in Kluska et al 2021 ; all traits in Cesarani, Masuda, et al 2021 ) and simulated populations ( Bradford et al 2019 ). Other studies ( Macedo, Christensen, et al 2020 ; Cesarani, Biffani, et al 2021 ) have reported dispersion of various models to be mostly within 10% of the expected value (1.0), which was also our result.…”

Genomic Selection Using Single-Step Genomic BLUP on the Number of Services per Conception Trait in Thai–Holstein Crossbreeds

“…Misztal et al [20] reported that the accuracy of the GEBVs was greater than that of the pedigree-based EBVs. At the same time, Sungkhareecha et al [43] reported that the rate of genetic progress from the top 20% of the herd found that the ssGBLUP method was faster than the traditional BLUP method in milk traits, and it would be helpful for increasing the selection accuracy and reducing the costs of progeny testing. Although the accuracy of the GEBVs was already sufficiently high for dairy cattle, the adoption of the technology has been widespread, and future work will focus on improving this accuracy.…”