Quantifying the Drivers of Genetic Change in Plant Breeding

Hybrid breeding programmes exploit heterosis to improve genetic gain. In particular, Maize exhibits high levels of heterosis. Since Shull (1908) formulated the concept of hybrid varieties in the early 20th century, the traditional open pollinated varieties were progressively replaced by hybrids from the cross between complementary lines belonging to different heterotic pools. Although it has been long recognized that efficient genetic diversity is required across pools, there has been little research into how the germplasm exchange between pools contributes to the genetic mean and variance over time. This research aims to partition the additive genetic values given the heterotic pool and inbred/hybrid germplasm origin into contributions to the mean and variance trends summarised over the years. A single trait representing grain yield controlled by 3,000 QTLs uniformly distributed over ten chromosomes is simulated. Each QTL is assigned an additive and a dominance genetic effect. Two germplasm exchange scenarios were defined as i) 0% and ii) 40% of 40 elite yield trial germplasm imported into the breeding programme yearly from an external source. We also investigated additive and dominance genetic by environment (GxE) interaction by simulating multiple locations. Three scenarios for GxE interaction were then considered: i) no GxE interaction and genetic correlation between imported and breeding germplasm equal to one; ii) same as i) but genetic correlation equal 0.7; and iii) with GxE interaction and a complex genetic correlation structure between imported and breeding germplasm. It is also common to suppose the environment has a possible interaction effect with year. Thus, we have simulated the main effects of year and location and their interaction effect. The results showed that we could track the contribution of the imported germplasm to the genetic mean and variance over time. Furthermore, we show how this contribution differs for each heterotic pool, with or without GxE interaction, and under different genetic correlations between the breeding programme and the external source.