Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Animal Genomics and Improvement Laboratory » Research » Publications at this Location » Publication #382570

Research Project: Improving Dairy Animals by Increasing Accuracy of Genomic Prediction, Evaluating New Traits, and Redefining Selection Goals

Location: Animal Genomics and Improvement Laboratory

Title: Improving the model for genetic evaluation of calving traits in the US Holstein and Brown Swiss

item BIFFANI, STEFANO - Collaborator
item TIEZZI, FRANCESCO - North Carolina State University
item DURR, JOAO - Council On Dairy Cattle Breeding
item COLE, JOHN - Former ARS Employee
item Vanraden, Paul
item MALTECCA, CHRISTIAN - North Carolina State University
item NICOLAZZI, EZEQUIEL - Council On Dairy Cattle Breeding

Submitted to: Interbull Annual Meeting Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 3/15/2021
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Many factors can cause under- or over-estimation of breeding values, such as parental selection, preferential treatment, selective reporting, and data manipulation. Even in the genomic era, serious concerns arise on the preselection of young bulls based on their genomic breeding values estimated by both a multiple-step procedure or by a single-step genomic prediction approach. Nevertheless, an additional and possibly underestimated source of bias can be simply related to excluding one source of variation from the model used for the analysis. For instance, in dairy cattle, preadjusting milk production for age and parity can cause serious under- or over-estimation of the genetic trend. Some authors reported that redefining the herd effect in the Finnish repeatability animal model evaluation reduced bias considerably. Moreover, ICAR’s guidelines on Dairy Cattle Genetic Evaluation stress the importance of model’s unbiasedness, especially in the framework of international genetic evaluation. Data from the official April 2019 run were used to investigate possible improvements of the current National Genetic evaluation for Calving Ease (CE) and Stillbirth (SB). Currently, CE and SB are evaluated separately using a Single-trait threshold sire-maternal grandsire (MGS) model. The model includes the following environmental effects: Random herd-year (HY), fixed year-season, parity-sex, sire/mgs birth year group, MGS breed (CE only). Breeding values are expressed as percent difficult births (score 4 & 5) observed in first calf heifers and percent stillbirths (score 2 & 3) observed over all parities for CE and SB, respectively. Preliminary analyses suggested a set of possible model improvements: exclusion of any herds with more than 95% of easy calvings, use of category 4 for category 4 and 5, the inclusion of parity in the definition of random HY groups, the inclusion of the interaction of Parity-Sex-Year of birth of Sire and MGS. Model improvements were successfully validated by ITB methods 1 and 3 and implemented in 2020.