Genomic prediction of threshability in naked barley. Issue 2 (6th February 2023)

• Record Type: Journal Article
• Authors: Massman, Chris; Meints, Brigid; Hernandez, Javier; Kunze, Karl; Smith, Kevin P.; Sorrells, Mark E.; Hayes, Patrick M.; Gutierrez, Lucia
• Is Part Of: Crop science. Volume 63:Issue 2(2023)
• Page Start: 674
• Page End: 689
• Publication Date: 2023-02-06
• Subjects: Crop science -- Periodicals; Cultures -- Périodiques; Cultures de plein champ -- Périodiques; Crop science; Nutzpflanzen; Zeitschrift; Pflanzenbau; Periodicals; 633
• DOI: 10.1002/csc2.20907
• Languages: English
• ISSNs: 0011-183X

Abstract

Abstract: Threshability, defined here as the propensity of grains to lose their hull after harvest, is a key trait in naked barley ( Hordeum vulgare L.). While threshability is a defining characteristic of naked grains and has been found to be associated with grain size and shape, its genetic architecture is poorly described. The goals of this study were to identify quantitative trait loci associated with threshability and evaluate their utility as covariates in genomic prediction models. A genome‐wide association study identified two loci on chromosomes 2H and 3H associated with threshability. The locus on chromosome 2H accounted for 9.9% of the phenotypic variance explained (PVE). The locus on chromosomes 3H accounted for 7.8% of the PVE. With effects on threshability of 0.18 and 0.29 for each marker, respectively, these markers could have a limited impact when implemented in marker‐assisted selection. Predictive ability for threshability was 0.842 using a structured genomic best linear unbiased prediction model. Incorporation of the markers with significant associations as covariates in the model did not improve predictive ability. Predictive ability was improved by the use of a multi‐trait model including grain test weight. The high predictive ability for threshability overall indicate that genomic selection would be useful in selection for threshability in naked barley. Core Ideas: Two markers were identified with significant associations to threshability. Predictive ability for threshability was 0.842 using a structured gBLUP model. Predictive ability was not increased by inclusion of identified QTL as covariates. Predictive ability was increased using a multi‐trait prediction model.