Genomic selection (GS) uses genomewide molecular markers to predict breeding values and make selections of individuals or breeding lines prior to phenotyping. Here we show that genotyping-by-sequencing (GBS) can be used for de novo genotyping of breeding panels and to develop accurate GS models, even for the large, complex, and polyploid wheat (Triticum aestivum L.) genome. With GBS we discovered 41,371 single nucleotide polymorphisms (SNPs) in a set of 254 advanced breeding lines from CIMMYT's semiarid wheat breeding program. Four different methods were evaluated for imputing missing marker scores in this set of unmapped markers, including random forest regression and a newly developed multivariate-normal expectation-maximization algorithm, which gave more accurate imputation than heterozygous or mean imputation at the marker level, although no signifi cant differences were observed in the accuracy of genomic-estimated breeding values (GEBVs) among imputation methods. Genomic-estimated breeding value prediction accuracies with GBS were 0.28 to 0.45 for grain yield, an improvement of 0.1 to 0.2 over an established marker platform for wheat. Genotyping-bysequencing combines marker discovery and genotyping of large populations, making it an excellent marker platform for breeding applications even in the absence of a reference genome sequence or previous polymorphism discovery. In addition, the fl exibility and low cost of GBS make this an ideal approach for genomics-assisted breeding.

Theoretical considerations suggest that wheat yield potential could be increased by up to 50% through the genetic improvement of radiation use efficiency (RUE). However, to achieve agronomic impacts, structural and reproductive aspects of the crop must be improved in parallel. A Wheat Yield Consortium (WYC) has been convened that fosters linkage between ongoing research platforms in order to develop a cohesive portfolio of activities that will maximize the probability of impact in farmers' fields. Attempts to increase RUE will focus on improving the performance and regulation of Rubisco, introduction of C(4)-like traits such as CO(2)-concentrating mechanisms, improvement of light interception, and improvement of photosynthesis at the spike and whole canopy levels. For extra photo-assimilates to translate into increased grain yield, reproductive aspects of growth must be tailored to a range of agro-ecosystems to ensure that stable expression of a high harvest index (HI) is achieved. Adequate partitioning among plant organs will be critical to achieve favourable expression of HI, and to ensure that plants with heavier grain have strong enough stems and roots to avoid lodging. Trait-based hybridization strategies will aim to achieve their simultaneous expression in elite agronomic backgrounds, and wide crossing will be employed to augment genetic diversity where needed; for example, to introduce traits for improving RUE from wild species or C(4) crops. Genomic selection approaches will be employed, especially for difficult-to-phenotype traits. Genome-wide selection will be evaluated and is likely to complement crossing of complex but complementary traits by identifying favourable allele combinations among progeny. Products will be delivered to national wheat programmes worldwide via well-established international nursery systems and are expected to make a significant contribution to global food security.

In genome-enabled prediction, parametric, semi-parametric, and non-parametric regression models have been used. This study assessed the predictive ability of linear and non-linear models using dense molecular markers. The linear models were linear on marker effects and included the Bayesian LASSO, Bayesian ridge regression, Bayes A, and Bayes B. The non-linear models (this refers to non-linearity on markers) were reproducing kernel Hilbert space (RKHS) regression, Bayesian regularized neural networks (BRNN), and radial basis function neural networks (RBFNN). These statistical models were compared using 306 elite wheat lines from CIMMYT genotyped with 1717 diversity array technology (DArT) markers and two traits, days to heading (DTH) and grain yield (GY), measured in each of 12 environments. It was found that the three non-linear models had better overall prediction accuracy than the linear regression specification. Results showed a consistent superiority of RKHS and RBFNN over the Bayesian LASSO, Bayesian ridge regression, Bayes A, and Bayes B models.

T ë~gënëtic yield progress of 26 spring wheat {Triticum aestivum L.

Conceptual models of drought-adaptive traits have been used in breeding to accumulate complementary physiological traits (PT) in selected progeny, resulting in distribution of advanced lines to rain-fed environments worldwide by the International Maize and Wheat Improvement Center (CIMMYT). Key steps in PT breeding at CIMMYT include characterisation of crossing block lines for stress adaptive mechanisms, strategic crossing among parents that encompass as many target traits as possible and early generation selection (EGS) of bulks for canopy temperature (CT). The approach has been successful using both elite × elite crosses as well as three way crosses involving stress adapted landraces. Other EGS techniques that are amenable to high throughput include measurement of spectral reflectance indices and stomatal aperture-related traits. Their genetic-and cost-effectiveness are supported by realisation of genetic yield gains in response to trait selection, and by economic analysis, respectively. Continual reselection within restricted gene pools is likely to lead to diminishing returns, however, exotic parents can be used to introduce new allelic diversity. Examples include landraces from the primary gene pool, and products of inter-specific hybridisation with the secondary gene pool consisting of closely related wheat genomes. Both approaches have been successful in introducing stress-adaptive traits. The main problem with knowing which genetic resource to use in wide-crossing is the uncertainty with which phenotypic expression can be extrapolated from one genome/genepool to another because of their unimproved or undomesticated genetic backgrounds. Nonetheless, their PT expression can be measured and used as a basis for investing in crossing or wide crossing. Discovering the genetic basis of PT is highly complex because putative QTLs may interact with environment and genetic background, including genes of major effect. Detection of QTLs was improved in mapping populations where flowering time was controlled, while new mapping populations have been designed by screening potential parents that do not contrast in the Rht, Ppd and Vrn alleles. Association genetics mapping is another approach that can be employed for gene discovery using exclusively agronomically improved material, thereby minimising the probability of identifying yield QTLs whose alleles have been already improved by conventional breeding.