作者
Hans D Daetwyler, Mulusew Fikere, Denise M Barbulescu, Michelle M Malmberg, Fan Shi, Joshua CO Koh, Sally Norton, Phillip A Salisbury, Surya Kant, Pankaj Maharian, Joe Panozzo, German C Spangenberg, Noel OI Cogan
发表日期
2018/1/1
期刊
AusCanola 2018 Co-hosts
页码范围
54
简介
Genomic selection (GS) uses genome-wide markers, usually single nucleotide polymorphisms (SNP), genotyped on germplasm that also has field trial data to predict genomic estimated breeding values (GEBV) for lines that have only been genotyped. A diverse set of B. napus spring and winter lines was compiled which included pre-breeding lines, commercial varieties, and germplasm sourced from the Australian Grains Genebank. Field trials were carried out across three years. Blackleg disease resistance field trials with 600 spring and winter lines were planted with two replicates near Green Lake and Wickliffe Victoria in 2015. For 200 spring lines, agronomic and seed oil traits field trials were completed with two to three replicates per line near Horsham (irrigated) and Mininera (rainfed) in 2016, as well near Horsham (irrigated and rainfed) in 2017. Internal infection and emergence were recorded in each year, whereas other traits may only be available for subsets of years. The following traits were recorded: emergence, survival, internal blackleg infection, vigour, days to flowering, days to maturity, plant height, seed yield, seed moisture content. Seed composition was determined with NIR for seed oil and protein%, total seed glucosinolate, and the main fatty acids.
All lines were genotyped with an in-house transcriptome genotype-by-sequencing (GBS) protocol that resulted in up to 98,000 SNP after stringent quality control. A genomic relationship matrix was calculated using all quality-controlled SNP. Genomic best linear unbiased prediction (GBLUP) and BayesR GS methods were used. Genomic prediction accuracies were moderate to high for …