Evaluating selection of a quantitative trait: snow mold tolerance in winter wheat

Authors: KRUSE, ERIKA - Washington State University; Esvelt Klos, Kathy; MARSHALL, JULIET - University Of Idaho; MURRAY, TIMOTHY - Washington State University; Ward, Brian; CARTER, AARON - Washington State University

Submitted to: Agrosystems, Geosciences & Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/25/2019
Publication Date: 10/1/2019
Citation: Kruse, E.B., Esvelt Klos, K.L., Marshall, J.M., Murray, T.D., Ward, B.P., Carter, A.H. 2019. Evaluating selection of a quantitative trait: snow mold tolerance in winter wheat. Agrosystems, Geosciences & Environment. 2(1):1-8. https://doi.org/10.2134/age2019.07.0059.
DOI: https://doi.org/10.2134/age2019.07.0059

Interpretive Summary: Molecular genetic markers can be used to rapidly select wheat lines with improved disease resistance. Several methods are available for this purpose, including marker-assisted selection (MAS) and variations on genomic selection, the success of which depend on their ability to accurately model plant disease response under varying environmental conditions. The development of snow mold disease in wheat is heavily influenced by the environment, and tolerance to this disease in wheat is controlled by numerous genes each with a small effect. This presents a challenge for the use of genetic markers in selection for tolerance. This paper compares marker-assisted and genomic selection-based methods for their ability to predict the snow mold tolerance of wheat lines derived from a cross between cultivars Xerpha (susceptible) and Munstertaler (resistant). MAS and genomic selection methods both failed to predict snow mold tolerance in the Xerpha by Munstertaler cross. This is in contrast to the successful performance of genomic selection in a cross between Finch (resistant) and Eltan (susceptible) scored for snow mold tolerance under the same environmental conditions. Marker-assisted selection is unlikely to facilitate breeding efforts for highly quantitative traits such as snow mold tolerance, whereas genomic selection is promising, given sufficient replication across environments.

Technical Abstract: Core Ideas Six quantitative trait loci for snow mold tolerance were detected in a winter wheat recombinant inbred line population. Marker-assisted selection to incorporate QTL is unlikely to help breed for highly quantitative traits. Genomic selection could replace initial phenotyping for quantitative traits. Selection for snow mold tolerance (SMT) in winter wheat (Triticum aestivum L.) is complicated by the influence of numerous quantitative trait loci (QTL) and of environmental conditions. The goals of this study were to identify QTL for SMT, determine the effectiveness of marker-assisted selection (MAS), and model the effectiveness of genomic prediction for SMT. Quantitative trait loci analysis of a recombinant inbred line (RIL) subpopulation, derived from a cross between Xerpha and Münstertaler, detected six unique QTL. Progeny from the same cross were advanced by MAS and compared with the unselected subpopulation to evaluate the efficacy of MAS. No significant difference was found between the SMT means (p = 0.41). Similarly, genomic selection had very poor accuracy (-0.07) in the Xerpha–Münstertaler (XM) RIL subpopulation. This contrasts with the apparent effectiveness of genomic selection (0.65) in a Finch–Eltan RIL population, also evaluated for SMT. The failure of selection tools to improve SMT in the XM population is likely due to the challenges of rating a quantitative trait that requires highly specific environmental conditions for phenotype development.