Crafting for a better MAGIC: systematic design and test for Multiparental Advanced Generation Inter-Cross population.

Multiparental Advanced Generation Inter-Cross (MAGIC) populations are valuable crop resources with a wide array of research uses including genetic mapping of complex traits, management of genetic resources and breeding of new varieties. Multiple founders are crossed to create a rich mosaic of highly recombined founder genomes in the MAGIC recombinant inbred lines (RILs). Many variations of MAGIC population designs exist; however, a large proportion of the currently available populations have been created empirically and based on similar designs. In our evaluations of five MAGIC populations, we found that the choice of designs has a large impact on the recombination landscape in the RILs. The most popular design used in many MAGIC populations has been shown to have a bias in recombinant haplotypes and low level of unique recombinant haplotypes, and therefore is not recommended. To address this problem and provide a remedy for the future, we have developed the "magicdesign" R package for creating and testing any MAGIC population design via simulation. A Shiny app version of the package is available as well. Our "magicdesign" package provides a unifying tool and a framework for creativity and innovation in MAGIC population designs. For example, using this package, we demonstrate that MAGIC population designs can be found which are very effective in creating haplotype diversity without the requirement for very large crossing programs. Furthermore, we show that interspersing cycles of crossing with cycles of selfing is effective in increasing haplotype diversity. These approaches are applicable in species that are hard to cross or in which resources are limited.

Spatial variation in the degradation rate of the pesticides isoproturon, azoxystrobin and diflufenican in soil and its relationship with chemical and microbial properties.

The extent of within field variability in the degradation rate of the pesticides isoproturon, azoxystrobin and diflufenican, and the role of intrinsic soil factors and technical errors in contributing to the variability, was investigated in sites on sandy-loam and clay-loam. At each site, 40 topsoil samples were taken from a 160 x 60 m area, and pesticides applied in the laboratory. Time to 25% dissipation (DT25) ranged between 13 and 61 weeks for diflufenican, 5.6 and 17.2 weeks for azoxystrobin, and 0.3 and 12.5 weeks for isoproturon. Variability in DT25 was higher in the sandy-loam in which there was also greatest variability in soil chemical and microbial properties. Technical error associated with pesticide extraction, analysis and lack of model fit during derivation of DT25 accounted for between 5.3 and 25.8% of the variability for isoproturon and azoxystrobin, but could account for almost all the variability for diflufenican. Azoxystrobin DT25, sorption and pH were significantly correlated.