2010 Annual Report
1a.Objectives (from AD-416)
Genomic information, particularly information about DNA sequence polymorphism, has great potential to increase the rate of improvement in small grains breeding programs. As genotyping costs fall and genotyping services are made available through the USDA small grains genotyping labs, breeders will need new methods to apply those resources effectively to their selection programs. The overall objectives of this program are to develop effective methods for identification of quantitative trait loci (QTL) and marker-assisted selection (MAS), and to deliver those methods to breeders and geneticists in publications and software.
1: Develop methods for identifying plant breeding quantitative trait loci.
2: Integrate methods for QTL identification into strategies that enable geneticists and breeders to design more efficient experiments and make better selection decisions.
3: Develop breeder-friendly tools for genomic and genetic data access and analysis, with a specific focus on optimum analysis and use of molecular marker and agronomic data for small grains breeders and geneticists.
1b.Approach (from AD-416)
Objective 1: Develop methods for identifying plant breeding quantitative trait loci.
Experimental Design. Populations under a Wright-Fisher neutral model will be simulated using a standard coalescent approach using different parameter values to compare three analysis methods: 1. Single-marker regression, 2. A random-effect haplotype method, and 3. A coalescent-based haplotype method. These methods will be applied using different haplotype block identification methods.
Objective 2: Integrate methods for QTL identification into strategies that enable geneticists and breeders to design more efficient experiments and make better selection decisions.
Experimental Design. To predict specific untested haplotype-environment effects, the covariance matrix of haplotype-within-environment effects will be modeled in two ways. First, the covariance of haplotype main effects can be modeled on the basis of the sequence similarity of the haplotypes. Second, the covariance of haplotype effects across environments can be modeled much as the covariance of genotype effects in multi-environment trials can be modeled. We will also explore a combination of these two options. Simulations of MAS will be applied to data from the Barley CAP for spring, six-row barley. The form of the distribution of QTL effects obtained from the real data will be maintained. Mixed model and whole-genome selection methods will be applied.
Objective 3: Develop breeder-friendly tools for genomic and genetic data access and analysis, with a specific focus on optimum analysis and use of molecular marker and agronomic data for small grains breeders and geneticists.
Experimental Design. In collaboration with GrainGenes, displays currently available in TASSEL and Haploview will be scoped, resource requirements estimated, and priorities established. In addition, this project will provide association analyses based on diversity data stored in the GrainGenes database, with significant markers to be displayed on a genetic map. Methods developed in the preceding two objectives will be implemented as plugins to the TASSEL software package. TASSEL already handles most of the data input, data management, and output functions. Connections will be established between GrainGenes, The Hordeum Toolbox (THT), and USDA small grains genotyping labs by implementing a GDPC (Genomic Diversity and Phenotype Connection) web-service for each database.
We continued work on basic methodology for using high-density DNA markers to accelerate crop improvement. This work included assessing the value of using sets of adjacent markers (so-called haplotype blocks) rather than single-markers alone, comparing genomic selection to marker-assisted selection, and developing methods to improve the performance of genomic selection for long-term gain. This work went in parallel with its application to two crops, barley and oat.
Work on breeding methods using DNA markers: We developed models to compare gains per unit time at fixed annual cost of genomic selection (GS) versus a standard marker-assisted selection (MAS) scheme. We found that the acceleration of the breeding cycle enabled by GS increases gain relative to MAS even when the accuracy of genomic predictions is lower than accuracies currently observed empirically.
Research on GS to date has shown that it can be effective for one or two cycles of selection, but what about the long-term? To address this question, we developed a programming environment to simulate multiple cycles of GS and other selection methods. We showed that GS can lead to a more rapid loss of favorable alleles than phenotypic selection but that the DNA marker data used by GS to make predictions can guide selection decisions to greatly reduce the loss of favorable alleles. A further effort of our lab has been to gather and review publications on genomic selection, particularly from animal breeding journals, and make them accessible for plant breeders. We have written two reviews of genomic selection for plant scientists.
Work on barley: Within the Barley Coordinated Agricultural Project (CAP) breeding lines we have systematically identified blocks of correlated adjacent markers (haplotype blocks). We evaluated the power of these methods in association mapping analyses. On simulated data, we found that haplotype methods were rarely beneficial. On real data, however, haplotype methods identified two validated genes affecting the trait that were missed by single-marker methods.
We published a simulation study to investigate in detail optimal methods using single markers to map QTL in the Barley CAP breeding lines. Two methods were evaluated to account for the complex genetic relationships among the lines and we identified one that had slightly but consistently higher power than the other. We also showed that for initial scans to identify genes affecting a trait, it can be useful to use a population of breeding lines that are more highly related, rather than highly diverse.
Work on oat: DNA marker data from lines from the last decade of oat breeding in North America have been analyzed to identify regions of the oat genome that harbor genes associated with beta-glucan content. Breeding programs for high beta-glucan content using these regions continues as a first cycle of selection will be completed in August 2010. DNA marker scores at 400 loci from a global oat collection have been analyzed to assess population structure and linkage disequilibrium within this crop.
Genomic selection increases crop improvement per unit time. Empirical estimates of the accuracy with which breeding line performance can be predicted using genome-wide markers are beginning to be reported. These accuracies are generally lower than the accuracy possible from actually phenotyping a line. But phenotyping takes time, slowing down the breeding cycle relative to genomic selection. ARS scientists at the Robert W. Holley Center for Agriculture & Health in Ithaca, NY developed theoretical models of selection and coupled them to a simple economic accounting of the cost of different operations (e.g., phenotyping, genotyping, and inbreeding) to analyze the annual gain from phenotypic versus genomic selection under equal costs. We found that the accuracy required of genomic predictions was surprisingly low, lower than the typical empirical estimates available now. This analysis suggests that genomic selection should already be competitive with phenotypic selection and the barriers to adoption are primarily those of transition costs.
Marker information used for genomic selection can also serve to maintain genetic diversity. Theoretical and empirical predictions of genomic selection have, to date, addressed short-term gains limited to one or two cycles of selection. The question of how well genomic selection might perform over the long term has not been addressed. ARS scientists at the Robert W. Holley Center for Agriculture & Health in Ithaca, NY addressed this question using simulation, which allow us to simulate twenty cycles of selection. We showed initially that while genomic selection enables rapid initial gains, favorable alleles are also lost from the population leading to a lower selection limit for genomic than phenotypic selection. Subsequently, we developed a method capitalizing on the marker data used for genomic predictions to also maintain allelic diversity and reduce the loss of favorable alleles from a population under selection. Using this method, genomic selection out-performed phenotypic selection in both the short and long terms. There exists now substantial interest in the adoption of genomic selection among applied plant breeders and researchers. Our research provides practical methods to implement genomic selection that substantially improve its long-term benefit without reducing rapid initial gains.
Multi-marker methods for identifying genes. Genome-wide association studies seek to identify genes that affect important but complex traits. These methods consist of correlating variants in DNA markers with phenotypes: high correlation indicates that the DNA marker is close to a causal gene. To date, these methods have primarily used single-marker methods. It is also possible, however, to combine multiple adjacent markers into so-called haplotypes and to correlate them with the phenotype. ARS scientists at the Robert W. Holley Center for Agriculture & Health in Ithaca, NY explored the value of this approach using simulation and on real data. In simulations, we found that overall, single markers did as well or better than haplotype approaches. On real data, however, we identified two genes using haplotypes that were not identified using single markers. Thus, the developed multi-marker methods may provide a more complete description of genes.
Tinker, N.A., Kilian, A., Wright, C.P., Heller-Uszynska, K., Wenzl, P., Rines, H.W., Bjornstad, A., Howarth, C., Jannink, J., Anderson, J.M., Rossnagle, B.G., Stuthman, D.D., Sorrells, M.E., Jackson, E.W., Tuvesson, S., Kolb, F.L., Olsson, O., Federizzi, L.C., Carson, M.L., Ohm, H.W., Molnar, S.J., Scoles, G.J., Eckstein, P.E., Bonman, J.M., Ceplitis, A., Langdon, T. 2009. New DArT markers for oat provide enhanced map coverage and global germplasm characterization. BMC Medical Genetics. 10:39.
Hamblin, M.T., Close, T.J., Bhat, P.R., Chao, S., Abraham, K., Blake, T., Brooks, W.S., Cooper, B., Griffey, C.A., Hayes, P.M., Hole, D.J., Horsley, R.D., Obert, D.E., Smith, K.P., Ullrich, S.E., Muehlbauer, G.J., Jannink, J. 2010. Population structure and linkage disequilibrium in US barley germplasm: implications for association mapping. Crop Science. 50:556-566.