Source: AGRICULTURAL RESEARCH SERVICE submitted to
ASSOCIATION GENETICS OF BETA-GLUCAN METABOLISM TO ENHANCE OAT AND BARLEY GERMPLASM FOR FOOD AND NUTRITIONAL FUNCTION
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
NRI COMPETITIVE GRANT
Reporting Frequency
Annual
Accession No.
0211227
Grant No.
2008-55301-18746
Project No.
NYR-2007-01682
Proposal No.
2007-01682
Multistate No.
(N/A)
Program Code
56.0A
Project Start Date
Dec 15, 2007
Project End Date
Dec 14, 2011
Grant Year
2008
Project Director
Jannink, J. L.
Recipient Organization
AGRICULTURAL RESEARCH SERVICE
(N/A)
ITHACA,NY 14853
Performing Department
US PLANT SOIL & NUTRITION LAB
Non Technical Summary
This integrated project addresses CSREES Goal 4 of improving the Nation's nutrition and health. It supports Goals 1) by providing economic opportunity to oat producers, 3) by increasing food security by boosting domestic food supply and training plant breeders, and 5) by increasing demand for oat, a sustainable crop that can reduce pest problems and chemical use. The activities proposed address three emerging trends: the distancing of the public from the practice of plant breeding; increasing attention to the genetics of a crop's nutritional value; and the advent of marker systems and analyses enabling population-wide association between markers and alleles that improve valuable traits. Our long-term goals are to enhance the security and nutritional value of grain by 1) contributing to knowledge needed to improve grain for nutrition, 2) devising breeding methods that optimally use biotechnology, and 3) developing education for the next generation of plant breeders. The objectives of this proposal are to 1) identify loci affecting health-promoting β-glucan through association genetics on elite oat and use these loci to compare competing marker-assisted selection (MAS) methods, 2) identify loci affecting β-glucan in oat from the National Plant Germplasm System (NPGS) and determine whether NPGS alleles can complement elite germplasm, 3) use genomic resources in barley to identify loci affecting β-glucan and develop oat markers for them, and 4) teach professionals association-based MAS and draw intelligent students toward plant breeding as a rewarding career.
Animal Health Component
(N/A)
Research Effort Categories
Basic
75%
Applied
25%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
2011560108050%
2021560108125%
2041560108025%
Knowledge Area
202 - Plant Genetic Resources; 204 - Plant Product Quality and Utility (Preharvest); 201 - Plant Genome, Genetics, and Genetic Mechanisms;

Subject Of Investigation
1560 - Oats;

Field Of Science
1080 - Genetics; 1081 - Breeding;
Goals / Objectives
Objective 1. Association mapping of β-glucan content and in elite oat germplasm and replicated comparison of phenotypic with marker-assisted selection. We will apply association analysis to a population of elite oat lines tested in the USDA UOPN since 1996, represented by about 410 lines of known pedigree. The β-glucan content of lines is known; we will also measure the molecular weight of the polymer. We will genotype lines using DArT markers developed through a consortium of 15 oat research groups. We will apply both mixed-model association analysis controlling for kinship using pedigree information and whole genome analysis. Results will feed into two cycles of MAS using both analyses. Conversion of UOPN data files on GrainGenes to a relational database that will also hold genotypic data will provide informatic support and a future resource for the oat community. Objective 2. Association mapping of β-glucan content from the National Plant Germplasm System: complementation of elite oat. The GRIN system contains β-glucan content observations on 5382 oat accessions. We will pick 200 lines from each tail of the distribution to genotype with the marker system described above. Their kinship matrix will be constructed using genotypic data. The tails of the distribution differ by 3% in β-glucan content and this selective genotyping will confer substantial power to the analysis. Objective 2 will either confirm loci identified in Objective 1, or to identify new loci. For loci mapped in Objectives 2 but not 1, we will determine whether the elite population carries the allele conferring higher β-glucan content as shown by analysis of the NPGS population. This Objective will identify not only accessions but alleles from the NPGS oat collection useful for oat improvement. Objective 3. Educational initiatives to pipeline students into plant breeding and to educate professionals. We will develop an interactive distance short course. The course syllabus and content will emerge from the research proposed here and from the barley CAP. The target audience will be plant breeding professionals and upper-level graduate students in plant breeding nationwide. Two educational activities will further help draw new students to plant breeding. First, a one-hour presentation oriented toward high-school students will be used at recruitment venues organized by ISU. Second, a week-long teaching module on biotechnology and computational biology in plant breeding will be introduced into two lower-level Agronomy courses and one upper-level Food Science course to show plant breeding to be an exciting and rewarding area for graduate study and professionally. These aims will serve our stakeholders by attracting quality students to plant breeding and by providing continuing education for their employees.
Project Methods
OBJECTIVE 1. Association mapping of β-glucan content in elite oat germplasm and comparison of phenotypic with marker-assisted selection. Aim 1.1. Association analysis in North American elite oat. Entries submitted to the UOPN over the past 12 years will be genotyped with DArT markers. We will use known pedigrees and markers to calculate kinships. We will study linkage disequilibrium decay with genetic distance. We will perform mixed-model association analysis and apply whole-genome analyses. Aim 1.2. Management of oat phenotypic and DArT genotype data. Currently the data from the UOPN is available online as spreadsheets. This information will be migrated to a relational database holding both it and the DArT marker data from Aim 1.1. Aim 1.3. Compare three selection methods. Phenotypic selection. We will reanalyze all UOPN data jointly using mixed-model analysis and employ rapid cycle recurrent selection to recombine them. Mixed-model MAS. After association analysis described in Aim 1.1. all loci identified as contributing to β-glucan content will be entered into a single statistical model for prediction of genotypic value. Whole genome selection. Here, all markers are retained in the model, and the shrunken genetic effects of all markers are estimated. Comparison of selection methods. All three selection schemes will go through two cycles of selection. Each program will be replicated twice. Evaluation will allow a comparison of methods. OBJECTIVE 2. Association mapping of β-glucan content from the National Plant Germplasm System: complementation of elite oat. Aim 2.1. Association analysis in NPGS oat. Using GRIN we will identify divergent germplasm. Single plants representative of chosen accessions will be genotyped. Aim 2.2. Compare association in the NPGS population with those in elite oat. Loci that are associated in the NPGS but not in elite oat will provide a list of candidate loci for introgression from NPGS accessions to elite oat lines. OBJECTIVE 3. Educational initiatives to pipeline students into plant breeding and to educate professionals. Aim 3.1. Recruiting potential students to the study of plant breeding. High school level. We will develop presentations that inform potential students about careers in plant breeding and the impact plant breeding research has on the larger community, using the example of breeding oats and barley with enhanced β-glucan. Our program will target outreach toward women and minority students. Undergraduate level. We will create curriculum focusing on the science of plant breeding and the impact of plant breeding programs, on the economy and on health. Curriculum will be inserted into introductory undergraduate courses taught at Iowa State University. Aim 3.2. Education of professionals, advanced plant breeding students, and undergraduate students. We will develop a graduate credit course on both theoretical and practical aspects of applying DNA marker information to breeding. The course will be accessed by computer and deliver course content in an interactive multimedia.

Progress 12/15/07 to 12/14/11

Outputs
Target Audience: Small grains breeders and geneticists. Particularly oat breeders and other grain breeders concerned with grain quality. Crop biochemists may also be interested in our results. The educational initiatives target primarily established breeders who were not trained in breeding methods using DNA marker data. Changes/Problems: The main change to report was in Objective 3: Educational initiatives to pipeline students into plant breeding and to educate professionals. We had intended to develop a distance education short course. Difficulties in coordination due to the lead PI's departure from Iowa State University caused us to change this so that the PhD students Mark Newell and Franco Asoro developed curricula that are now being integrated into a distance master's program at Iowa State University as chapters in three courses: Molecular Plant Breeding, Molecular Genetics and Biotechnology, Quantitative Genetics for Plant Breeding. All three courses are delivered through on-campus and online resources (http://agonline.iastate.edu/programs/master-science-plant-breeding-distance-option). What opportunities for training and professional development has the project provided? The project funded two PhD students who successfully completed their degrees. How have the results been disseminated to communities of interest? Besides the six publications in key crop breeding journals, results have been disseminated through invited presentations at scientific meetings (three presentations at the Plant and Animal Genome meetings in 2009, 2011, & 2013, in San Diego, CA, and one presentation at the Crop Science Society meetings in 2011 in San Antonio, TX.). Mark Newell (PhD candidate) also gave four presentations to perspective students at Iowa State University and participated in three science-fair community activities on behalf of the project (summarized here: http://gfspopgen.agron.iastate.edu/GFSPopGen_outreach.html) What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? The grant resulted in the completion of Five major activities. 1. Markers are useful because they segregate with causal polymorphisms, so-called quantitative trait loci. Prior to using markers for breeding, therefore, it is essential to know how they may be associated with QTL by calculating linkage disequilibrium among markers. This paramter quantifies the non-randomness of association between loci in the genome. The first major activity, therefore, was a comprehensive study of linkage disequilibrium in oat for a global dataset. 2. We then moved to using the markers to identify loci affecting the important quality trait, beta-glucan content. Identification was performed in two distinct populations. One a global population obtained from the national small grains collection, and the other a North American population of elite oat obtained from cooperative breeding evaluations run by the US Department of Agriculture. 3. For the purpose of oat improvement, identifying loci that affect the phenotype is useful, but it is also useful to simply use the markers to generate more accurate predictions of the performance of new breeding lines. This type of prediction is called genomic selection. Another major activity was a feasibility study of genomic selection using the marker datat that we had on the elite oat population. In that study we evaluated the accuracy with which line performance could be predicted based on past breeding data and genotypes of new breeding lines. 4. The final key component of scientific research was to roll this previous research together into a selection program. We implemented three selection methods (phenotypic, marker-assisted, and genomic selection), each replicated twice. Two cycles of selection were completed, with each cycle taking one year. The selection goal was high beta-glucan content, with some effort to maintain genetic variability. 5. The fifth activity was educational. Two Ph.D. students were trained under this grant. Both students prepared curricula for course modules on using association genetics in plant breeding, one set on genome wide association studies and the other on genomic selection. These curricula are being used in distance education courses at Iowa State University. Significant results from the research were: 1. Oat can be subdivided into a number of sub-populations. These stratify oat by production region and adaptation. 2. Linkage disequilibrium in cultivated oat is high. This means that despite oat's large genome, only a few thousand markers are needed to cover it sufficiently. At the time of the research, the best marker option available to the oat community was DArT markers, which generally yielded a number of polymorphic markers in the elite population close to the lower limit of what was needed. Recent activity in sequencing, however, has generated marker systems that provide ample markers. 3. A number of promising candidate loci for high beta-glucan content were identified. Functional candidates for the loci that were associated with the phenotype were identified by sequence similarity to rice. These results provide a number of targets for marker-assisted selection. In addition, we found that at least one prominent locus that was polymorphic in the global population from the National Small Grains Collection was not polymorphic in the elite oat population, suggesting an opportunity for introgressing an allele confering high beta-glucan content from exotic sources. 4. Two excellent PhD candidates were mentored through their terminal degrees. One has continued in academia, working for the Noble Foundation in Ardmore, OK as their rye breeder, and the other working as a breeder for Monsanto.

Publications

  • Type: Journal Articles Status: Published Year Published: 2012 Citation: Newell, M.A., F.G. Asoro, M.P. Scott, P.J. White, W.D. Beavis, and J.-L. Jannink. 2012. Genome-wide association study for oat (Avena sativa L.) beta-glucan concentration using germplasm of worldwide origin. Theor. Appl. Genet. 125:1687-1696.
  • Type: Journal Articles Status: Published Year Published: 2013 Citation: Asoro, F.G., M.A. Newell, M.P. Scott, W.D. Beavis, and J.-L. Jannink. 2013. Genomewide Association Study for Beta-glucan Concentration in North American Elite Oat. Crop Sci. 53: 542-553
  • Type: Journal Articles Status: Awaiting Publication Year Published: 2013 Citation: Newell, M., D. Cook, H. Hofmann, and J.-L. Jannink. 2013. An Algorithm for Deciding the Number of Clusters and Validating using Simulated Data with Application to Exploring Crop Population Structure. The Annals of Applied Statistics. In Press.
  • Type: Journal Articles Status: Awaiting Publication Year Published: 2013 Citation: Asoro, F., M. Newell, W. Beavis, P. Scott, N. Tinker, and J.-L. Jannink. 2013. Comparison of Genomic, Marker-Assisted, and Pedigree-BLUP Selection Methods to Increase ?-Glucan Concentration in Elite Oat Germplasm. Crop Science In Press.


Progress 12/15/07 to 12/14/08

Outputs
OUTPUTS: Objective: A) identify loci affecting health-promoting β-glucan through association genetics on elite oat and use these loci to compare competing breeding methods 1. A Ph.D. student (Franco Asoro) has been hired to lead this project. 2. Seed from 446 elite spring oat lines that have been evaluated over the past 12 years in the Uniform Oat Performance Nursery were obtained; DNA extracted from single plants of each line; DArT markers scored; Redundancy analysis performed on the markers, yielding 1005 markers with unique profiles across the germplasm. 3. β-glucan content data from the UOPN were analyzed using mixed model methods to obtain BLUP of β-glucan content for all lines. X loci were found significantly associated to β-glucan content in this data set, using a Y false discovery rate. The range of phenotypic variance explained by the QTL was . 4. These same oat lines were planted in field experiments in Ames, IA in 2009 and 2010. We are processing all these samples to obtain improved estimates of β-glucan content in each line before analyzing these results for publication. We anticipate publishing in Fall 2011. 5. Six breeding programs for high β-glucan content were initiated as described in previous progress reports. We have two replicated programs for each of phenotypic selection, marker-assisted selection, and genomic selection. We have completed the second selection event in all programs and made crosses among selections. Progeny from these crosses will be planted in the field in spring 2011 for comparison of progress among the programs. Objective: B) Identify loci affecting β-glucan in oat from the National Plant Germplasm System (NPGS) and determine whether NPGS alleles can complement elite germplasm Accomplishments: 1. A Ph.D. student (Mark Newell) has been hired to lead this project. 2. Germplasm from the NPGS divergent for β-glucan content has been identified and grown in the field in Ames, IA. In 2009 and 2010. 3. Marker data from these accessions has been obtained. 4. We are currently processing samples from field evaluations to measure β-glucan content. 5. We will initiate association analysis once we have all the data. We anticipate publishing in Fall 2011. Objective: C) Teach professionals association-based MAS and draw students toward plant breeding as a rewarding career. 1. Since last year we have developed outlines for four modules that will be a part of a distance course delivered through Iowa State University's planned distance master's degree in plant breeding. The modules are 1. Populations genetics: allele frequencies and linkage disequilibrium; 2. Association mapping in breeding programs; 3. Association-based marker-assisted selection; and 4. Genomic selection. These modules will be integrated into a course called molecular plant breeding that will be delivered in Spring 2012. PARTICIPANTS: Training: Mark Newell, Ph.D. candidate. Leads the project to identify loci contributing to β-glucan content from NPGS accessions. Also has developed and given presentations to high-schoolers and in introductory agronomy courses on opportunities and science within modern plant breeding Franco Asoro, Ph.D. candidate. Leads the project to identify loci contributing to β-glucan content in elite sprint accessions, and to compare phenotypic, marker-assisted, and genomic selection methods. Franco performed association analyses presented at PAG. Collaborations: This project has deepened collaborations with the oat group at Agriculture and Agri-food Canada headed by Dr. Nick Tinker. TARGET AUDIENCES: Plant breeding professionals who need further education in how high-throughtput markers are changing plant breeding. Accomplishments: 1. Since last year we have developed outlines for four modules that will be a part of a distance course delivered through Iowa State University's planned distance master's degree in plant breeding. The modules are 1. Populations genetics: allele frequencies and linkage disequilibrium; 2. Association mapping in breeding programs; 3. Association-based marker-assisted selection; and 4. Genomic selection. These modules will be integrated into a course called molecular plant breeding that will be delivered in Spring 2012. PROJECT MODIFICATIONS: Not relevant to this project.

Impacts
The two publications produced so far are at the cutting edge of population genetic analysis in oat and of empirical genomic selection evaluation in the public sector in general. They are helping in bringing these ideas forward for practical small grains breeding programs.

Publications

  • 1. Newell, M.A., D. Cook, N.A. Tinker, J.-L. Jannink. 2010. Population structure and linkage disequilibrium in oat (Avena sativa L.): Implications for genome-wide association studies. Theor. Appl. Genet. 122:623-632. 2. Asoro, F.G., M.A. Newell, W.D. Beavis, M.P. Scott and J.-L. Jannink. 2011. Accuracy and training population design for genomic selection in elite north american oat (Avena sativa L.). The Plant Genome, in review.