한우 도체형질 육종가 신뢰도 추정을 위한 PEV와 MTEDC 방법의 비교 연구

This study estimated the reliability of EBV (Estimated Breeding Value) used for animal improvement by PEV (Prediction Error Variance) and MTEDC (Multiple-trait Effective Daughter Contribution) methods respectively. The reliability of Hanwoo carcass traits EBV was evaluated and compared. The reliability of EBV was analyzed based on data extracted from random sampling by iterating 10 times at 1,391,141 animals. The reliability validation methods are compared by PEV and MTEDC. In Data 1 of PEV, the reliability of CW (Carcass Weight), EMA (Eye Muscle Area), BF (Backfat Thickness) and MS (Marbling Score) was estimated using PEV and the reliability of each traits was analyzed as 0.38, 0.39, 0.40 and 0.42. In Data 2 of MTEDC, the reliability was estimated using MTEDC and the reliability of each trait was analyzed as 0.45, 0.48, 0.48 and 0.52. The estimated reliability of the cows progeny animals (0-5) of CW was 0.30-0.64 in Data 1 and 0.42-0.52 in Data 2. The reliability of the EMA was 0.57-0.67 in Data 1 and 0.47-0.55 in Data 2, the reliability of BF was 0.31-0.67 in Data 1 and 0.43-0.55 in Data 2 and the reliability of MS was estimated to be 0.33-0.73 in Data 1 and 0.43-0.60 in Data 2. It is difficult to estimate a reliability of 0.5 or higher due to the low progeny of Hanwoo cows. Accordingly, it is estimated that more than 100 animals related to pedigree are needed to improve reliability. To this end, it is considered important that the number of offspring of each sire is included in the evaluation of more than 100 records in the data composition for estimated breeding value and reliability. It is suggested that the nationwide pedigree data and performance test data, which can implement robust relationship matrices in the data configuration for EBV evaluation, should be analyzed together.