Novel studies into the improvement of cattle fertility using multiple technologies that converges bioinformatics with machine learning.

Abstract

This study aimed to evaluate the presence of a panel of 18 candidate lethal recessives in the Irish dairy, and beef cattle populations. Establishing the presence and frequency of such recessive alleles will assist the scientific, and animal breeding communities. The frequency of variants in Irish Holstein – Friesian dairy cattle are included in the dairy chapter. Furthermore, the beef cattle chapter assesses six different breeds Aberdeen Angus, Belgian Blue, Charolais, Hereford, Limousin and Simmental. Animal breeders have established that the optimal breeding goal is a more balanced and informed breeding approach; thus, the lethal recessives in the various population were assessed for any associated effects on 16 production traits in both the dairy and beef cattle prior to any breeding decisions made on-farm, for example culling, which may reduce the overall genetic diversity, increasing inbreeding depression and ultimately reducing population viability. A comparative genomics approach followed the above candidate lethal recessive association studies. This chapter identified variants in essential genes, using alternative mammalian species studies as supporting evidence. Variants were identified and prepared for genotyping approximately 350 K cattle, the results were analysed for potential candidates associated with embryo loss. Furthermore, variants within a 500 kb range of the candidates were selected and submitted to identify segregating variants that were possibly unknown to industry. Genetics and high throughput data are one aspect of improving cattle fertility. It is clear that in addition to genomic programmes, predictive technologies will be used in making forecasts in the agricultural sector, particularly machine learning and AI models. Such predictive technologies may ultimately incorporate genomic and environmental data into selection programmes; however, interdisciplinary studies are currently at the point of combining technologies. ReproDoc assesses cattle pregnancy status using mobile ultrasound technology and has amassed a database of animal-related details. This chapter prepared a private commercial database for predictive models to determine when individual cows are optimal for insemination. Customers would be the first to benefit from the success of such a technology, as the requirement for semen, and insemination attempts would decrease, and the need for herd replacements would be optimised, ultimately reducing the demand for replacements. Such information could provide farmers reassurance in allowing a cow to remain on the farm until the next breeding cycle, thus improving welfare and reducing wastage.