Investigating alternative strategies to evaluate the feasibility of genomic selection for heat tolerance in US Holstein cattle

Authors: FRAGOMENI, BRENO - University Of Connecticut; DUTTA, GAURAV - University Of Connecticut; SCHOBER, HENRY - Collaborator; MCWHORTER, TAYLOR - Council On Dairy Cattle Breeding; BIFFANI, STEFANO - Collaborator; TIEZZI, FRANCESCO - University Of Florence; Miles, Asha

Submitted to: Plant and Animal Genome Conference Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 11/8/2024
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Heat stress represents a major concern for the United States dairy industry and contributes to significant economic losses every year. Many studies in the past two decades have demonstrated that genetics plays an essential role in heat stress tolerance and a breeding program could serve as a tool to mitigate its effects. Additionally, studies have demonstrated that there is genetic variation for heat stress tolerance in the US Holstein population. However, there is still no national genomic evaluation for heat tolerance in the US. Therefore, the short-term goal of this study is to evaluate the feasibility of a national evaluation for heat stress in US Holstein cattle. The long-term goal of this study is to implement a national evaluation for heat tolerance in the US in partnership with the stakeholders. Data was provided by the Council on Dairy Cattle Breeding (CDCB) and by the Cooperative Dairy DNA Repository (CDDR) and included records of more than 10M cows born between 2000 and 2020. Pedigree information was available for more than 65M animals. Additionally, genotypes from all bulls with daughters in this dataset were available for 70K SNPs, totaling 46,044 genotypes. Access to this dataset was the result of a collective effort from the University of Connecticut, the CDCB, and the CDDR and allows the research team to work on the data without compromising data integrity and security. A test day model in a two-trait approach was suggested as a phased approach between the 305-day lactation yield and a random regression test day model. Initial analyses were performed for test day milk yield. Results collected on the state of Texas demonstrate that modeling test day records using a binary heat stress definition result is possible. The heritabilities for the test days under heat stress (0.25) were higher than the single trait test day (0.23), and the no-heat stress binary (0.20). Genetic correlations between records collected under heat stress and in thermoneutral conditions were 0.91, indicating modest sire reranking. Many analyses resulted in numeric problems or slow convergence due to the large number of observations and the complexity of the statistical model. Computational burden is expected to increase once data from other states is analyzed together. Therefore, the next steps of this project will focus on efficient implementations that will allow simpler models and reduced data to predict heat tolerance for PTA using the national data. The next steps of this project, which will assist with the long-term goals are to expand the model to evaluate the impact of heat stress in protein, fat and somatic cell count. Additionally, a genome-wide association study will be implemented to search for genomic regions associated with heat tolerance. Finally, our future goals include the evaluation of fertility and health traits and a trade-off analysis between heat tolerance and production.