Genomic selection in global diversity pools-a new paradigm for pre-breeding in canola

Abstract

We propose a new paradigm for pre-breeding in canola based on genomic selection in a globally diverse pool of <i>Brassica napus</i> germplasm. The base population comprising over 400 diverse canola lines representing global genetic diversity in the crop was developed at NSW DPI, Wagga Wagga. In this study, we assessed genetic diversity among 368 canola lines representing <u>A</u>ustralian <i><u>B</u>rassica <u>n</u>apus</i> <u>H</u>omozygous <u>D</u>iversity <u>S</u>et (ABnHDS) by Illumina SNP markers. These lines were evaluated for early vigour, flowering time, grain yield, and resistance to blackleg under field conditions as well as for tolerance to Mn²⁺ under controlled environment conditions. A subset of 196 lines was also analysed for structural variations (INDELS) and SNPs using whole genome resequencing approach. Cluster and population structure analyses of ABnHDS (368 lines) grouped canola lines based on their growth habit and geographic origin. Using both molecular and phenotypic data, individual lines were assigned genomic predicted breeding values (GPBV) by genomic best linear unbiased prediction (GBLUP). Genomic prediction accuracies ranged from 47 to 80% across traits (except for 37% for flowering time under short day conditions) indicating that genomic prediction can be used quite confidently to predict the performance for some canola lines with missing phenotypes. We further employed mating design based on optimal contributions selection (OCS) to manage long-term genetic gain and genetic diversity in the global diversity pool. Canola breeders will have the accelerated access of lines with potential value in their commercial programs based on their ranking for economic index and GPBV for individual traits. The global diversity pool will also be useful to identify the role of individual genes that underlie traits of economic importance.