Genetic variants deciphered to improve cattle health and productivity

Each bovine animal carries several million variations in its genome, known as genetic variants. By combining genetic data from cattle reared on French farms with measurements of their performance, it is possible to identify the variants associated with one or more biological characteristics. However, this statistical approach alone does not reveal whether these variants have a direct effect on genome function. For this, so-called “functional” approaches are required, aimed at understanding the direct impact of these variants on gene expression and function. This research has a dual impact: firstly, on a fundamental level, to understand how the bovine genome functions, and secondly, in terms of practical applications, as this information will ultimately enable the development of more precise and effective genomic selection tools. Thus, the researchers involved in this study asked themselves the following question: How do variants in the bovine genome influence the variability of major agronomic traits?

High-throughput functional analysis combining quantitative genetics, molecular screening and artificial intelligence

To achieve their goal, the researchers pooled their expertise in different fields of biology to develop an approach for analyzing a large number of variants in the bovine genome, and thus revealing those that impact biological traits. Statistical genetic analyses were first used to isolate a set of genomic variants potentially involved in modifying gene functionality. Then, a combination of bioinformatics methods based on artificial intelligence algorithms and high-throughput molecular screening implemented on cell models enabled us to validate the functional character of some of these variants.

Many genetic variants impact agronomic traits by disrupting the splicing mechanism.

This statistical selection approach, combined with functional validation, has enabled us to identify 38 genomic variants that modify the way genes are expressed and interpreted by the cell, with repercussions on bovine traits. It was specifically designed to identify variants that alter a molecular process known as splicing. Splicing occurs during gene expression and, by modifying messenger RNAs, modulates the quantity or sequence of proteins produced. The variants we have identified alter this mechanism, thereby modifying cell function. On a larger scale, this leads to macroscopic modifications whose effects can be observed in bovine agronomic traits. These include milk production (both quantity and composition), morphology, udder health and fertility. This is a major advance in our understanding of how the bovine genome works. It is true that variants impacting splicing had been little studied in this species, and this work has revealed that they play a predominant role in the development of traits.

Towards more agro-ecological cattle farming

By identifying variants with a direct biological effect on the levels of bovine traits, it is possible to better target the variants that should be taken into account for genomic selection to facilitate the implementation of a more agroecological agriculture. This work should contribute to the breeding of animals that are more productive, more robust and better adapted to the challenges of animal health and sustainability. It also demonstrates the power of approaches combining quantitative genetics, functional genomics and artificial intelligence to decipher the complexity of genome functioning.

Reference : Charles M, Gaiani N, Sanchez MP, Boussaha M, Hozé C, Boichard D, Rocha D, Boulling A. Functional impact of splicing variants in the elaboration of complex traits in cattle. Nat Commun. 2025 Apr 24;16(1):3893. doi: 10.1038/s41467-025-58970-5.