Durum wheat production is globally important, but grain yield has been stagnating in recent decades. In order to ensure that its production maintains the pace with increasing demand, breeding for high grain yield must be supported by molecular-based methods. Genomic estimated breeding values for selection and genome scan were assessed as molecular tools holding maximum potential for durum wheat breeding. Four recombinant inbred line populations bred by inter-mating elite were sown in yield trials at five sites. All progenies were characterized using "genotyping by sequencing" method. A consensus map was developed, and missing genotypes were imputed using a Hidden Markov model to reach a total of 1987 polymorphic markers. Models accounting for genotype environment interactions were used to estimate the genetic component of each measured trait. Hence, Bayesian ridge regression was used to determine the predicted values and their relative accuracy in several combinations, testing full-sibs and half-sibs as training population for grain yield and 1,000 kernel weight. The high level of accuracy achieved suggests that GEBV for selection holds great potential for durum wheat breeding, as long as full-sibs are used as training populations, in combination with statistical models that account for G?E. In order to test the exploitability of genome scan to guide breeding crosses, a separate genome-wide association study was conducted. 288 elite were sown in the south of Morocco and at two sites along the Senegal River for two years. These sites show a temperature differential of 10?C. Implementing a GE model facilitated identifying the most heat tolerant among the tested entries. 8,173 polymorphic SNPs were inquired, and several associations could be identified between markers and the ability to withstand the heat gradient. Hence, GWAS holds great potential to increase genetic gain in breeding via increased accuracy in determining the crosses to be made.
Displaying 1 - 6 of 6
The Global Wheat Program of CIMMYT is one of the largest public breeding programs in the world consisting of millions of lines/ genotypes derived from thousands of crosses evaluated under using a shuttle breeding cycle and multi-environment testing. The germplasm is phenotyped for conventional (such as yield and grain quality) as well as non-conventional traits (physiological traits) in field and greenhouse conditions. The breeding germplasm is also screened with genome-wide markers (using Illumina SNP array, genotyping-by-sequencing, or DArTseq platforms) and/or multiple gene/QTL region-specific molecular markers (using KASP platform). All genotyped samples are registered in the "DNA SampleTracker," a software system for tracking DNA samples developed at CIMMYT. In collaboration with High Throughput Genotyping Platform project, the plant sample and data collection methods are optimized. Meanwhile, the extensive wheat genealogies and phenotypic information have been maintained in the International Wheat Information System and will be transferred to a new Enterprise Breeding System. Furthermore, several bioinformatics/statistical genetics methods with the objectives of gene discovery and genomic prediction have been developed and utilized for optimizing genomics-assisted selection. The wheat team is a member of "Genomic Open-source Breeding Informatics Initiative (GOBII)" which aims to develop and implement genomic data management systems to enhance the capacity of breeding programs. Under this initiative, a new genomics database has been built and a pilot wheat version is being tested at CIMMYT. Several decision support tools are also under collaborative development, such as a Genomic Selection Pipeline based on Galaxy, Flapjack-based F1/line verification, and marker assisted backcrossing tools. Additional tools are envisioned for the future including a Cross-Assistor and Selection-Assistor. The ultimate aim is to seamlessly connect the genomic database, phenotypic database, and decision support tools to support the breeding selection process and to lead to the development of cultivars with increased rates of genetic gain.
Genomic selection facilitates rapid cycling through a breeding cycle by eliminating the need to phenotype prior to selecting superior parents and crossing among them. In winter wheat we can now complete a cycle of GS in about 12 months and two greenhouse seasons. Season consists of planting F1s from the previous cycle and selfing to obtain F2 seed. The second season involves planting and genotyping the F2s, predicting their value with GS, selecting and crossing the best, and harvesting the F1 seed. Our breeding program has completed five cycles of GS in one population primarily for grain yield, over the past five years. We have completed three cycles of GS for resistance to Fusarium Head Blight in a second population. Genotyping was done using genotyping-by-sequencing. This provides an opportunity to assess the changes in the population that have occurred as a result of this rapid cycling. These include 1) changes in genomic estimated breeding values for grain yield and FHB resistance, 2) effect of selection and drift on allele frequencies including fixation, 3) effect of selection on diversity and genetic relationships, and 4) changes in linkage disequilibrium. We are conducting these analyses and will present the results.
Durum wheat is the tenth most important crop in the world, but its cultivation is mostly limited to harsh, arid, and heat prone marginal lands. Breeding for tolerance to these conditions is often considered the most strategic approach to ensure adaptation, especially when paired with best agronomical practices. The word 'adaptation' summarizes all the research efforts conducted to identify the many traits controlling the mechanisms for withstanding or escaping the traceries of the environment. It can be summarized as "GGE vs E". The durum wheat breeding program of ICARDA deploys targeted phenotyping methods in combination with genomic scans to dissect these 'adaptive' traits into simple loci. These loci can then be pyramided via a combination of international field testing, markers assisted selection, genetically-driven crossing schemes, and genomic selection to derive climate-ready cultivars. Here, several examples of this approach are presented and their implications for 'adaptation' are discussed.
Stem rust is a globally important wheat disease that can cause severe yield loss. Breeding for quantitative stem rust resistance (QSRR) is important for developing cultivars with durable resistance. Genomic selection (GS) could increase rates of genetic gain for quantitative traits, but few experiments comparing GS and phenotypic selection (PS) have been conducted. Our objectives were to compare realized gain from GS based on markers only with that of PS for QSRR in spring wheat using equal selection intensities; determine if gains agree with theoretical expectations; and compare the impact of GS and PS on inbreeding, genetic variance, and correlated response for pseudo-black chaff (PBC), a correlated and likely pleiotropic trait. Over two years, two cycles of GS were performed in parallel with one cycle of PS, with each method replicated twice. For GS, markers were generated using genotyping-by-sequencing, the prediction model was initially trained using historical data, and the model was updated before the second GS cycle. Overall, GS and PS led to a 31±11 and 42±12% increase in QSRR and a 138±22 and 180±70% increase in PBC, respectively. Genetic gains were not significantly different, but were in agreement with expectations. Per year, gains from GS and PS were equal, but GS led to significantly lower genetic variance. This shows that while GS and PS can lead to equal rates of short-term gains, GS can reduce genetic variance more rapidly. Further work to develop efficient GS implementation strategies in spring wheat is warranted.
Trait and trait stability are important for wheat breeding. Our objectives were to assess the relative efficiency of genomic selection (GS) for various wheat traits and trait stability using four different models. Genotyping was conducted with a 90K SNP chip panel. SNP tagging was used to obtain a subset of 3,919 relatively independent markers for downstream analysis. Phenotyping was conducted on a population consisting of 273 lines, from seven different soft red winter wheat breeding programs in the U.S.A. Eberhart and Russells’ regression and additive main effects and multiplicative interaction (AMMI) models were used to assess trait stability. GS accuracy (r) was assessed for ridge regression best linear unbiased prediction (rr-BLUP), Bayesian ridge regression (BRR), reproducing kernel hilbert space (RKHS), and elastic net (EN). Across all models, GS produced a wide range of accuracies for trait stability (0.1 to 0.65) that varied by trait and stability method. Accuracy was 0.35 for yield and 0.44 for yield stability using AMMI, indicating the viability of GS in selecting lines with both high and stable yield. Our findings lay the foundation for wheat breeding programs in northeastern U.S.A. to implement GS. It also provides useful information for wheat rust researchers: as phenotypic selection for rust resistance can be both expensive and time consuming and rapid evolution of rust pathogens require an emphasis on durable resistance controlled by multiple genes, the GS approach applicable for complex traits thus has potential to achieve higher gains per unit time than traditional breeding for rust resistance.