Exploring test-day production yields for US Holstein cows in Texas state for heat-tolerance genomic predictions

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

Submitted to: Plant and Animal Genome Conference Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 11/7/2024
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: The effect of genotype by environment (GxE) interactions and heat stress on dairy cattle significantly influences milk production and quality, fertility, health, and overall animal well-being. Production data collected in different environments can be used to model effects of heat stress and understand how genetics respond to different environmental conditions using reaction norm models. The goal of this is to investigate the extent of heat stress using test-day milk yields in US Holstein cows in the state of Texas, as the initial step towards developing genomic predictions for heat tolerance. Data was collected from the Council in Dairy Cattle Breeding in the state of Texas from animals born after year 2000 and only the first lactation were considered. The variance component for milk yield was estimated using an Average Information Restricted Maximum likelihood algorithm. Initially, a 305-day milk yield model was fitted with 241,554 records from daughters 2,175 of bulls with least 30 progeny. This model mimics the national evaluation and is used as a benchmark for comparison of results. Following the preliminary analysis, 331,973 test-days for 49,626 Holstein obtained from the 10 largest herds in the state of Texas were fitted using a single trait repeatability test-day model. In a second group of analysis, data was divided into two categories based on months of the test day, and records were assigned to summer and no-summer. Two definitions of summer months were used: May to September and June to August. The heritability (SE) of test-day milk yield was estimated at 0.23 (0.01), which was lower than the 305-day milk yield analysis at 0.29 (0.01). The differences between those analyses are expected due to the differences and the model and in the animals used. For the test-day results, the permanent environment variance was similar to the additive genetic variance indicating the impact of nongenetic factors. In the two-trait approach, the June to August summer and non-summer months heritability was 0.20(0.01) and 0.25(0.01), respectively with a genetic correlation of 0.91(0.01). Initial attempts of the May to September summer definition did not converge. Those values indicate that HS can be modelled with a two-trait analysis and indicate a potential reranking of animals during the summer months. The next steps towards the implementation of a national evaluation are to calculate a heat load function using weather station data to split the data into records collected into hot or thermoneutral conditions. The heat load will later be used fit a random regression model. Additionally, data from other states will be incorporated, and genomic data will be used to calculate breeding values.