Submitted to: Theoretical and Applied Genetics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: July 15, 2012
Publication Date: August 7, 2012
Citation: Newell, M., Franco, A., Scott, M.P., White, P., Beavis, W., Jannink, J. 2012. Genome-wide association study for oat (Avena sativa L.) beta-glucan concentration using germplasm of worldwide origin. Theoretical and Applied Genetics. 125:1687-1696. Interpretive Summary: Beta glucans have been shown to promote cardio vascular health when eaten in sufficient quantity. One of the best sources of beta glucans is oats. It would be desirable to increase the level of beta glucans in oats and other foods. In order to do this, it is important to understand what genes are responsible for deposition of beta glucans. We therefore set out to identify genes that control beta glucan content in oat using a method called GWAS. Traditional methods of gene identification are difficult to apply in oats, but GWAS proved to be well-suited to this species. We identified three genes that control beta-glucan levels in oat. These genes can be used by breeders to increase the level of beta glucans in oats and other species, resulting in food products that support cardio vascular health.
Technical Abstract: Detection of quantitative trait loci (QTL) controlling complex traits followed by selection has become a common approach for selection in crop plants. The QTL are most often identified by linkage mapping using experimental F2, backcross, advanced inbred, or doubled haploid families. An alternative approach for QTL detection are genome-wide association studies (GWAS) that use pre-existing lines such as those found in breeding programs. We explored the implementation of GWAS in oat (Avena sativa L.) to identify QTL affecting b-glucan concentration, a soluble dietary fiber with several human health benefits when consumed as a whole grain. A total of 431 lines of worldwide origin were tested over two years and genotyped using Diversity Array Technology (DArT) markers. A mixed model approach was used where both population structure fixed effects and pair-wise kinship random effects were included. Various mixed models that differed with respect to population structure and kinship were tested for their ability to control for false positives. As expected, given the level of population structure previously described in oat, population structure did not play a large role in controlling for false positives. Three independent markers were significantly associated with b-glucan concentration. Significant marker sequences were compared with rice and one of the three showed sequence homology to genes localized on rice chromosome seven adjacent to the CslF gene family, known to have b-glucan synthase function. Results indicate that GWAS in oat can be a successful option for QTL detection, more so with future development of higher-density markers.