Population structure and genome-wide association analysis for frost tolerance in oat using continuous SNP array signal intensity ratios

Abstract

Key messageInfinium SNP data analysed as continuous intensity ratios enabled associating genotypic and phenotypic data from heterogeneous oat samples, showing that association mapping for frost tolerance is a feasible option.AbstractOat is sensitive to freezing temperatures, which restricts the cultivation of fall-sown or winter oats to regions with milder winters. Fall-sown oats have a longer growth cycle, mature earlier, and have a higher productivity than spring-sown oats, therefore improving frost tolerance is an important goal in oat breeding. Our aim was to test the effectiveness of a Genome-Wide Association Study (GWAS) for mapping QTLs related to frost tolerance, using an approach that tolerates continuously distributed signals from SNPs in bulked samples from heterogeneous accessions. A collection of 138 European oat accessions, including landraces, old and modern varieties from 27 countries was genotyped using the Infinium 6K SNP array. The SNP data were analyzed as continuous intensity ratios, rather than converting them into discrete values by genotype calling. PCA and Ward’s clustering of genetic similarities revealed the presence of two main groups of accessions, which roughly corresponded to Continental Europe and Mediterranean/Atlantic Europe, although a total of eight subgroups can be distinguished. The accessions were phenotyped for frost tolerance under controlled conditions by measuring fluorescence quantum yield of photosystem II after a freezing stress. GWAS were performed by a linear mixed model approach, comparing different corrections for population structure. All models detected three robust QTLs, two of which co-mapped with QTLs identified earlier in bi-parental mapping populations. The approach used in the present work shows that SNP array data of heterogeneous hexaploid oat samples can be successfully used to determine genetic similarities and to map associations to quantitative phenotypic traits.Electronic supplementary materialThe online version of this article (doi:10.1007/s00122-016-2734-y) contains supplementary material, which is available to authorized users.

Genome-wide association analysis for lodging tolerance and plant height in a diverse European hexaploid oat collection

“…Before calculation all missing values were replaced by the mean ratio per marker. The genetic structure of the collection is described in detail in Tumino et al (2016).…”

“…SNP data curation is described in detail in Tumino et al (2016). Briefly, SNP fluorescence signals were used to calculate intensity ratios according to the formula X/(X ?…”

“…This is particularly true for allopolyploids. For the 6K Oat Array, almost 20% of the polymorphic markers assayed in a diverse set of 595 mostly American oat varieties detected more than two clusters (Tinker et al 2014), which indicates hybridisation to homoeologous loci (Wang et al 2014), assuming accessions to be pure and nearly homozygous.…”

“…bund.de/aveq). Genetic diversity of the AVEQ collection and GWAS for frost tolerance were addressed by Tumino et al (2016) using continuous SNP intensity ratios, providing a framework for the present GWAS study. Objectives of this work are (1) to study the phenotypic variation of lodging and height in the AVEQ collection and their relationship, (2) to test the effectiveness of GWAS for finding loci associated with a complex trait such as lodging, (3) to provide new genetic markers to be used in breeding programs aimed to improve lodging tolerance in oat.…”

“…In the present study, we analysed lodging and plant height in a collection of European oat accessions including landraces, old and modern varieties, in order to perform a genome-wide association study (GWAS) for identifying markers associated to lodging tolerance. This collection has been recently genotyped by the Infinium 6K SNP array for oat and SNP data were analysed as continuous intensity ratios, rather than as discrete genotypes (Tumino et al 2016, Theor Appl Genet 129, pp. 1711-1724).…”