Balancing selection gain and genetic diversity in the genomic planning of crosses

Creating new genetic variation by crossing two or more parents is the initial and often most important step when developing new crop varieties. Hence, several mate selection indices have been suggested to support the planning of crosses in genomic breeding pipelines that have been established in many breeding programmes in recent years. The corresponding index weights are however difficult to determine objectively, and these indices often feature weights determined by laborious grid searches or rules of thumb. The aim of this study was to compare methods that employ the latter approach with mate selection indices based on desired gains in simulations and an empirical Fusarium head blight experiment for winter wheat. The results indicated that the suggested desired gain indices outperform routinely used methods in terms of reaching a favourable balance between the short-term selection gain, long-term selection gain, as well as the genetic diversity. They might have beyond that a high prospect for making a broader spectrum of genetic diversity accessible in the framework of germplasm exchange between breeding programmes.