Genomic selection for wheat traits and trait stability

Abstract

Based on the estimates of accuracy, genomic selection would be useful for selecting for improved trait values and trait stability for agronomic and quality traits in wheat. Trait values and trait stability estimated by two methods were generally independent indicating a breeder could select for both simultaneously. Genomic selection (GS) is a new marker-assisted selection tool for breeders to achieve higher genetic gain faster and cheaper. Breeders face challenges posed by genotype by environment interaction (GEI) pattern and selecting for trait stability. Obtaining trait stability is costly, as it requires data from multiple environments. There are few studies that evaluate the efficacy of GS for predicting trait stability. A soft winter wheat population of 273 lines was genotyped with 90 K single nucleotide polymorphism markers and phenotyped for four agronomic and seven quality traits. Additive main effect and multiplicative interaction (AMMI) model and Eberhart and Russell regression (ERR) were used to estimate trait stability. Significant GEI variation was observed and stable lines were identified for all traits in this study. The accuracy of GS ranged from 0.33 to 0.67 for most traits and trait stability. Accuracy of trait stability was greater than trait itself for yield (0.44 using AMMI versus 0.33) and heading date (0.65 using ERR versus 0.56). The opposite trend was observed for the other traits. GS did not predict the stability of the quality traits except for flour protein, lactic acid and softness equivalent. Significant GS accuracy for some trait stability indicated that stability was under genetic control for these traits. The magnitude of GS accuracies for all the traits and most of the trait stability index suggests the possibility of rapid selection for these trait and trait stability in wheat breeding.

Searching for an Accurate Marker-Based Prediction of an Individual Quantitative Trait in Molecular Plant Breeding

“…Further detailed simulations are possible to consider all existing marker prediction models, the related parameters associated with QTL genetic model, marker distribution and informativeness, training set and test environment (Heffner et al, 2009, 2010; Zhong et al, 2009; Hickey et al, 2014). However, our simulation results are consistent with several empirical reports from GS analyses that prediction accuracies were higher using only the QTL-linked markers or a subset of informative markers (e.g., Spindel et al, 2015; Thavamanikumar et al, 2015; Arruda et al, 2016; Edwards et al, 2016; Huang et al, 2016; Liu et al, 2016). Thus, the simulations in Angus cattle and soybean, along those empirical reports, provided support for our theoretical reasoning to search for more informative FAST SNP markers through RNA-Seq to improve trait prediction accuracy.…”

Association Analysis of Baking and Milling Quality Traits in an Elite Soft Red Winter Wheat Population

“…The phenotypes showed reasonably high heritability, as indicated in a prior genomic selection contribution by Huang et al (2016Huang et al ( , 2018 (Table 1). The phenotypes showed reasonably high heritability, as indicated in a prior genomic selection contribution by Huang et al (2016Huang et al ( , 2018 (Table 1).…”

“…The same germplasm was used by Huang et al (2016) to study genomic selection for trait stability. The panel was composed of lines from nine breeding programs serving nine states: Indiana, Michigan, Ohio, Missouri, Virginia, Kentucky, Maryland, Illinois, and New York.…”

“…We define year-location combinations as environments, which resulted in four environments denoted as OH2012, OH2013, VA2012, and VA2013. The same germplasm was used by Huang et al (2016) to study genomic selection for trait stability. A detailed description of experimental design and field experiments was given by Huang et al (2016).…”

Distinct genetic architectures for phenotype means and plasticities in Zea mays

“…Recent studies concluded that genomic prediction for yield stability in rye 9 and wheat 10 could be effective. However, to fully exploit plasticity it is important to understand its genetic architecture.…”