Abstract In recent years, genomic selection for wheat breeding has been widely studied, but this has typically been restricted to population sizes under 1000 individuals. To assess its efficacy in germplasm representative of commercial breeding programmes, we used a panel of 10,375 Australian wheat breeding lines to investigate the accuracy of genomic prediction for grain yield, physical grain quality and other physiological traits. To achieve this, the complete panel was phenotyped in a dedicated field trial and genotyped using a custom Axiom TM Affymetrix SNP array. A high-quality consensus map was also constructed, allowing the linkage disequilibrium present in the germplasm to be investigated. Using the complete SNP array, genomic prediction accuracies were found to be substantially higher than those previously observed in smaller populations and also more accurate compared to prediction approaches using a finite number of selected quantitative trait loci. Multi-trait genetic correlations were also assessed at an additive and residual genetic level, identifying a negative genetic correlation between grain yield and protein as well as a positive genetic correlation between grain size and test weight.
![Bread magic model 572 Bread magic model 572](/uploads/1/2/5/3/125396498/431707666.jpg)
Explore Heidi Abrams's board 'Salt dough crafts' on Pinterest.| See more ideas about Hobbyarbejde, Kreativ and Manualidades. Disturbed 665 dentures 453 breed 986 curiouser 561 behaviourist 758 slayin. Agentes 754 ealing 990 meadow 572 trails 358 lnyhbt 728 kimkardashian 28. 70 connection 203 buddhism 323 manualidades 583 largest 155 misogynist. 266 misuse 214 magic 782 dede 826 odio 3 hotdogs 491 alwayz 904 magia.
Plant breeding has been successful in producing significant yield gains in wheat since the beginning of the twentieth century (Wrigley and Rathjen ); this has largely been driven by the innovation and adoption of new breeding technologies. Such progress is underpinned by extensive research, first in developing the technology, and second on establishing its application. If new technologies are to continue enabling plant breeding to deliver genetic gain to growers, innovative research must be undertaken in datasets that are relevant to the setting in which they will be applied. Molecular markers are one technology that represent an invaluable research tool for understanding the genetic control of various traits. They have frequently been utilised in quantitative trait loci (QTL) mapping studies, and applied in breeding programmes through marker-assisted selection (MAS) (Koebner and Summers; Collard and Mackill ). Early statistical modelling approaches to QTL mapping involved the analysis of individual markers through simple scanning procedures (Soller et al. In more modern approaches, statistical methods have improved the efficiency and power of QTL detection through the simultaneous incorporation of markers from the whole genome in complex linear mixed models (Zhang et al.; Verbyla et al.
![Bread magic model 572 Bread magic model 572](/uploads/1/2/5/3/125396498/607806094.jpg)
There has also been focus on whole genome QTL mapping in broader multiparent populations (Huang et al.; Sannemann et al.; Mackay et al. ), and diverse association panels (Neumann et al.; Bentley et al.; Zanke et al. The latter usually involves the use of genome-wide association studies (GWAS) and has become a valuable tool for broad validation of previously identified QTL as well as identification of QTL in the target breeding germplasm. For qualitative traits under simple genetic control, GWAS, and subsequent application of MAS has been shown to be an effective tool in breeding programmes (Xu and Crouch ). However, for more complex polygenic quantitative traits such as grain yield, there have been few examples of genetic improvement using MAS (Dekkers et al.
This deficiency can be overcome by implementing a genomic selection (GS) method that uses a complete set of molecular marker effects for predicting the performance of quantitative polygenic traits (Meuwissen et al. Current research in this area suggests with sufficient prediction accuracy, GS can be successfully applied in a breeding programme to increase rates of genetic gain (Cooper et al.; Schmidt et al. Recent studies investigating the accuracy of GS in wheat have used population sizes ranging from several hundred to several thousand individuals, and achieved prediction accuracies mostly in the range of 0.50–0.60 as measured by Pearson correlation coefficients (Heslot et al.; Nakaya and Isobe; Isidro et al.; He et al. In GWAS and QTL analysis, the use of physical and genetic maps has been widely adopted (Kang et al.; Zhang et al. Recombination information from these maps could also be used in GS programmes to simulate the progeny of specific parents for the purpose of designing crosses (Podlich and Cooper ).
Physical maps are becoming available for wheat (Pozniak ), but can be of limited value if the individuals sequenced are not closely related to the target germplasm. Additionally, physical maps do not incorporate recombination information, which reduces their value when we are interested in simulating progeny based on recombination probabilities in the germplasm of interest.
Therefore, high-quality genetic maps built from relevant germplasm are a better resource for these applications. Examples of such maps in the literature include those produced using multi-parent advanced generation inter-cross (MAGIC) populations (Huang et al.; Gardner et al. ), as well as consensus maps constructed from multiple bi-parental populations (Cavanagh et al.; Wang et al. These maps can also be used to measure the extent of linkage disequilibrium (LD) between markers (Zhao et al.; Chao et al. In the context of association mapping and genomic prediction, LD becomes vitally important as it influences the achievable mapping resolution (Huang et al.
), power and accuracy of QTL detection (Somers et al. ), and the accuracy of genomic prediction in a breeding programme after multiple generations (Muir ). The extent of LD is also known to vary significantly depending on the germplasm structure (Hao et al.; Huang et al. ) and as a consequence, assessments of LD should be conducted on the genetic material being studied. For GS to be applied effectively, plant breeders must have a sound understanding of the relationship between traits of interest as it enables optimisation of selection strategies through correlated response to selection (Bernardo ).
Trait correlations in bread wheat have long been reported at the phenotypic level (Bhatt and Derera; Fischer and Wood ). Advances in statistical techniques have since made it possible to draw genetic correlations between traits by separating the genetic variance from the residual error (Gilmour et al. ), and these have been reported for various physiological traits in bread wheat (Rebetzke and Richards; Sukumaran et al. These approaches, coupled with the use of pedigree or molecular marker information, can also be used to separate the genetic variance into its additive and residual components, thus allowing genetic correlations to be drawn at the additive and residual genetic level (Rebetzke et al. These genetic correlations, particularly the additive, provide a more precise measure of trait relationships and facilitate better optimisation of selection strategies. GH PH LL LW Mat. NDVI TW TKW Yl.
Additive genetic correlations are in the upper triangle and residual genetic are in the lower triangle From each of the traits, the additive genetic values and residual genetic values were extracted from their respective fitted additive genotype models and used to understand genetic relationships between the traits. Table presents the pairwise additive and residual genetic correlations between traits analysed in the 2014 Roseworthy field trial.
The two correlation measures largely agreed, with a correlation of 0.79 across the 91 trait pairs. Of the 91 trait pairs, 74 had correlations in the same direction, and those that differed in direction were all near zero.
Additive genetic correlations were overall stronger than residual genetic with an absolute mean of 0.26 compared to 0.14. Notable correlations include the well-known strong negative relationship between grain yield and grain protein, with an additive correlation of −0.55 and a residual genetic of −0.30. A negative relationship was also observed between grain protein and test weight (additive correlation −0.22, residual genetic −0.43). Strong positive relationships were observed between test weight and thousand kernel weight (TKW) (additive correlation: 0.37, residual genetic 0.52), and relative maturity score and biomass (additive correlation 0.76, residual genetic 0.42).
A comparison of additive and baseline models. Broad sense heritabilities are presented for each model, and narrow sense for the genomic model as there is no term in the base model to capture the additive genetic variance. Model fit is compared through the log-likelihood measure a Proportion of the variance accounted for by the model that is additive All traits collected from the Roseworthy experiment were analysed and results from the fitted baseline models and additive genotype linear mixed models are compared in Table.
Additive models had significantly higher log-likelihood (model fit) for all traits, with an average improvement of 44% over the equivalent baseline models. The additive model also improved broad sense heritability for all traits, with an average increase of 24%. Narrow sense heritabilities of the additive models were comparable with the broad sense heritability from th.