Genome-wide association analyses reveal complex genetic architecture underlying natural variation for flowering time in canola

Optimum flowering time is the key to maximize canola production in order to meet global demand of vegetable oil, biodiesel and canola-meal. We reveal extensive variation in flowering time across diverse genotypes of canola under field, glasshouse and controlled environmental conditions. We conduct a genome-wide association study and identify 69 single nucleotide polymorphism (SNP) markers associated with flowering time, which are repeatedly detected across experiments. Several associated SNPs occur in clusters across the canola genome; seven of them were detected within 20Kb regions of a priori candidate genes; FLOWERING LOCUS T, FRUITFUL, FLOWERING LOCUS C, CONSTANS, FRIGIDA, PHYTOCHROME B and an additional five SNPs were localized within 14Kb of a previously identified quantitative trait loci for flowering time. Expression analyses showed that among FLC paralogs, BnFLC.A2 accounts for ~23% of natural variation in diverse accessions. Genome-wide association analysis for FLC expression levels mapped not only BnFLC.C2 but also other loci that contribute to variation in FLC expression. In addition to revealing the complex genetic architecture of flowering time variation, we demonstrate that the identified SNPs can be modelled to predict flowering time in diverse canola germplasm accurately and hence are suitable for genomic selection of adaptative traits in canola improvement programmes.