Dr Nelson Lubanga

Transition strategy for two-part genomic selection in plant breeding programmes

The project involves developing a transition strategy to enable plant breeding programmes transition from the current traditional approach to the two-part genomic selection strategy. The project is conducting stochastic simulations to evaluate the effectiveness of:

• A conventional approach to plant breeding;
• A conventional approach to the deployment of genomic selection in plant breeding;
• The two-part strategy;
• Various alternatives of the transition strategy.

The simulations are implemented using an R package AlphaSimR. The simulations consist of 20 years of a burn-in phase that will be shared by all strategies and an evaluation phase that simulate future breeding for another 20 years with each of the different breeding strategies. The strategies are compared in terms of genetic gain, genetic variance, and selection accuracy over time.

The results will guide the choice of the transition strategy for the canola breeding programmes of BASF. An implementation period will follow, when initial training sets will be assembled for selection of the parents to be used in the first population improvement cycle. Practical constraints to the modelled implementation can be empirically identified and the transition strategy optimised and adapted to real-world breeding programmes. As individuals progress from the population improvement component to the product development component, the performance of the two-part strategy can then be monitored in terms of genetic gain over time by successive field trial evaluations of selected genotypes.

This TRAIN@Ed project has received funding from the DDI programme, the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 801215, and industry funding from BASF.