SusAn 29 ReDiverse Biodiversity within and between European Red dairy breeds - conservation through utilization

The overall objective of the Norwegian research in REDIVERSE is to improve genomic prediction (GP) methods for across breed evaluation and for heterogeneous populations, while maintaining genetic diversity (GD). This is important because not all breeds involved in REDIVERSE are numerically large enough or lack the financial resources to establish a sufficiently large breed-specific reference population that enables GP with a competitive level of prediction accuracy. Especially for small breeds the number of bulls with daughter based breeding values to be included in the reference population is often limiting. This limitation can be relaxed by adding cows of the same breed to the reference population, or by adding bulls and cows from a different but related breed. Both solutions increase the heterogeneity of the phenotypes and genotypes used in the analyses. The Norwegian research efforts will develop procedures and tools to design reference populations that optimally use the heterogeneous data in GP. Phenotypic and genotypic data was collected on 60,000 Norwegian Red Cattle (NRF), 2500 Dutch MRY cattle, and 700 German RDN cattle. The genotypes of the NRF cattle contained ~700,000 markers, and those of the MRY and RDN cattle ~50,000 markers. After comparing genotypes across the breeds we found a set of 45204 SNPs that the breeds had in common and could be used for across breed analyses. The cleaning-up of the animals resulted in a set of 7248 Norwegian Red Cattle, 2507 Dutch MRY cattle and 801 German RDN cattle that are ready for analysis. An across breed genomic relationship matrix for all these animals and using the SNPs has been set up. Across breed genomic predictions have been performed, and accuracy was evaluated in an evaluation data set, that was kept separate from the data set used for the estimation of SNP effects. The Dutch data did have prediction accuracy to predict breeding values of German cows, and vice versa. However, the Norwegian data did not yield accurate predictions for German or Dutch cows, and vice versa. This was also expected from our estimates of the effective number of segments in the genome, which was very large across Dutch and German cows on the one hand and Norwegian on the other. The latter indicates that prediction across Dutch/German and Norwegian cows requires accurate prediction of many small chromosomal segments, which is hard in data sets of limited size.

The effective number of chromosomal segments turned out instrumental in deciding whether the joining of data sets across countries is beneficial for genomic prediction or not. If the effective number of segments across the breeds is of the same order of magnitude as within the breeds, combining of the data sets is helpful. If the nuber of segments between the breeds is one or more orders of magnitude higher than within the breeds, combining of the data is not helpful when using the genomic prediction methods used here. Possibly, with higher density SNP chips and variable selection genomic prediction methods, it may be possible to improve genomic prediction accuracies even if the effective number of segments is large.

Red dairy breeds across Europe represent a unique source of genetic diversity and are partly organized in trans-national breeding programs but are also well adapted to local conditions providing regional identity of products for consumers. The objective of REDIVERSE is to develop and to set in place collaborative and integrated novel breeding and management concepts to achieve a resilient and competitive use of these resources. A further goal is to strengthen best practices for small farmholders for improving product quality and to supply ecosystem services according to their specific circumstances. The challenge of establishing appropriate breeding and maintenance strategies for diverse farm systems and regional markets is met by multi-actor operations also considering economic, structural and social diversity in participating countries to offer tailored solutions. The holistic approach relies on integrative research of scientists in the fields of animal genetics, proteomics, economy and social sciences. Cutting edge technology such as large scale genomic and proteomic tools will be implemented to enhance genetic progress and to characterize specific properties. Innovative survey approaches will assess the impact of the sector on social acceptance and the needs of farmers. The project will generate novel knowledge and concepts that will be timely disseminated to lead-users such as the breeding and dairy industry, food sector, farmer cooperatives and farmers. A consistent monitoring of the consequences on the different levels when applying new tools or concepts is the key to balance the additional benefit to the sector on the various actors. The trade-off between economic, environmental and social interests will ensure sustainable dairy production, improve animal welfare and help to develop rural landscape.