8.10.5. MiXBLUP 3.0 (2024)

During the last decades, the mixed model equations have been used for estimating breeding values in livestock using phenotypes at the traits of interest, together with pedigree. With the introduction of genomic selection, the traditional mixed model has been extended to analyse simultaneously phenotypic, pedigree, and genomic information. Such models are commonly called single-step genomic models. Currently, genomic information used in routine evaluations contains the genotypes of several (hundred) thousand animals for around 50 thousand or more single nucleotide polymorphisms (SNP). The integration of such genomic datasets with large numbers of genotyped animals led to new challenges for solving the mixed model equations. Therefore, several approaches for solving the single-step models have been proposed in the literature and implemented in software for large genomic evaluations in livestock (e.g. for milk production in dairy cattle). For example, the software package MiXBLUP 3.0 allows to efficiently estimate genomic enhanced breeding values in livestock using various linear mixed effect models. MiXBLUP aims to be user-friendly while supporting efficient algorithms for solving large genomic evaluations. Originally developed for pedigree-based models, MiXBLUP has been extended to four different approaches for single-step genomic evaluations, allowing simultaneous analyses of all information for both genotyped and ungenotyped individuals.

A big advantage of this method is that all available data of relatives, genomic and phenotypic stored in large databases containing information of different sources and from different generations is used. This gives the opportunity to select not only for production traits, but simultaneously also for fitness and welfare traits. In this way selection of parents contributes indeed to a balanced breeding goal.