Product Line 1.2. Characterizing genetic diversity and creating novel gene pools
Rice diversity is the foundation for rice improvement programs. Intelligent use of this diversity can both help solve current production problems and create opportunities for future rice production and respond to climate change. To use fully the wealth of rice diversity, two essential ingredients are needed: we need to have the genetic blueprints of diverse rice germplasm accessions and varieties in use, and we need to generate plant populations with numerous recombined genotypes to allow full expression of phenotypic variation in order to discover new genes or QTLs for use in breeding programs. Rapid advances in DNA sequencing with declining costs will allow us to decode the genomes of a large number of rice accessions. This product line capitalizes on new sequencing, SNP genotyping, and phenotyping technologies to fully explore rice diversity while also creating populations suitable for trait dissection and the discovery of gene functions.
Two approaches will be taken to enable the efficient use of rice diversity: (1) develop a genetic diversity platform to predict gene-phenotype relationships, and (2) generate genetic stocks for trait dissection. First, we will create a SNP haplotype map to decode molecular variation. With public- and private-sector partners in the Rice SNP Consortium, at least 2,400 diverse accessions from Asia, Africa, and Latin America will be genotyped by a high-density SNP detection platform (950K Affymetrix chip) at a resolution of 1 SNP/kb of the rice genome. Purified seed stocks will be available to trait-expert partners in a Global Phenotyping Network who will apply lab and field-based methodologies for large-scale trait evaluation relevant to yield and climate-change-related stresses. We will capitalize on Cirad’s expertise and the investment made on developing methods, modeling tools, and high-throughput precision phenotyping for association genetics. Investments will be made to upgrade phenotyping capacity and develop high-throughput screening procedures for selected traits. The large data sets from genotyping and phenotypic evaluation will enable inference of important genotype-phenotype relationships by association genetics.
While the initial haplotype map will employ SNP chips, a longer-term goal is to sequence most of the rice germplasm collection once sequencing technologies deliver a low enough cost per accession. We will begin preparing a comprehensive DNA bank for at least 100,000 rice accessions using a high-throughput pipeline (2011-14). Low-cost sequencing techniques will be tested and optimized to determine the cost per accession (2012-13), followed by sequencing of 100,000 accessions (from 2014 onward). This comprehensive sequence information, together with geographical data on their origins and the trait associations detected in the initial set of 2,400 lines, will allow us to select specific accessions for evaluation of traits and isolation of novel genes.
For the second approach, we will produce specialized genetic stocks and populations that are rich in allelic diversity and genotypic combinations. We will generate diversity panels that include mutants of specific genotypic backgrounds, recombinant inbred lines (RILs), chromosomal segment substitution lines (CSSLs), and near-isogenic lines (NILs). Special emphasis will be given to the multiparent advanced generation intercrossed (MAGIC) technique with which repeated genetic recombination among multiple parents breaks down linkages to create novel genotypic combinations. RILs, MAGIC, and CSSLs will be developed with diverse indica, japonica, aus, aromatic, and Asian and African/Latin American AA-genome types that exhibit abiotic/biotic stress tolerance, wide adaptation, high yield potential, and good grain quality for use in breeding. Special efforts for the African rice, Oryza glaberrima, will include the development of intraspecific and interspecific materials and deep sequencing of the parental set used to construct CSSLs and iBridges lines. These multipurpose populations will enable precise QTL estimation and gene identification for both the cultivated and wild rice gene pools.
Because of the large amount of data produced, a bioinformatics pipeline for data analysis and documentation is essential. We will create a database to manage and visualize rice genetic and phenotypic data from thousands of genomes that are linked to IRIS GRIMS along with a toolkit to analyze these data. Existing technology and information from public open-source databases and software solutions will be adapted, and then deployed to meet the specific needs of rice. This will be done in partnership with ARI partners that have expertise in high-throughput data analysis and management, and comparative biology, so that existing tools and resources are effectively cross-linked to those developed by GRiSP. For instance, we will build on the database architecture analysis tool developed by Cirad for functional analysis (OryGenesDB and OryzaTagLine), for comparative genomics between Arabidopsis thaliana and rice genomes (GreenPhylDB). User-centric workbenches that merge gene function evidence from heterogeneous distributed rice resources will be developed.
1.2.1. High-resolution SNP genotypes of diverse accessions (Rice SNP Consortium)
1.2.2. A global phenotyping network for key agronomic traits and responses to major stresses, including climate-change traits
1.2.3. Whole-genome sequencing of all unique germplasm accessions held in the genebanks at IRRI, AfricaRice, and CIAT and other genetic stocks
1.2.4. Specialized genetic stocks and novel populations through enhanced recombination of cultivated and wild rice gene pools
Product 1.2.1 is primarily developed by the Rice SNP Consortium, which currently includes Cornell University, IRRI, Cirad, AfricaRice, CIAT, USDA, Syngenta, Bayer CropScience, and the Taiwan Agricultural Research Institute. We expect additional partners to join this consortium in the next two years. The consortium will be responsible for the production of high-density genotypes for a large collection of germplasm nominated by members. Product 1.2.2 on a global phenotyping network will include most of the members of the Rice SNP Consortium but also additional partners with specialized expertise in phenotyping. These include Colorado State University, JIRCAS, the China National Rice Research Institute (CNRRI), and interested NARES partners in IRRI breeding networks. Because of the diverse needs in trait evaluation, we expect that the phenotyping network will continue to grow over the course of the project. Product 1.2.3 on genebank sequencing will involve the best technology providers for low-cost sequencing. Currently, we are in discussion with the Beijing Genomics Institute (BGI).
For product 1.2.4 on the development of specialized genetic stocks, our partners include IRD, IRRI, AfricaRice, Cirad, CIAT, USDA, Cornell University, and selected NARES. These institutions have a long history of producing novel genetic resources and disseminating them for public use.
New genes and alleles for adaptive traits, including traits related to climate change, identified through association analysis are the main deliverables resulting from the various activities in this product line. These products are of a global nature with tremendous scientific and practical significance. The immediate users will be rice breeders and geneticists in advanced research institutes, NARES, and the private sector. The knowledge and novel genetic resources produced will accelerate the varietal development objectives in theme 2, thus delivering improved varieties to farmers and consumers at a faster pace throughout the world. In the near term, applications of the genetic diversity platform will concentrate on traits that are particularly relevant to expected changes associated with global climate change such as the occurrence of extreme weather and extreme temperatures, with too much or too little water, and more severe epidemics of diseases and insect pests. For the long term, the genetic diversity platform will have broad applications to infer relationships between genomic variation and phenotypic diversity of rice in multiple traits. Across MPs, the data sets on genotype and phenotype will be closely linked to the Genomic Integrated Breeding Service coordinated by the Generation Challenge Program.
Because of the broad interest in the use of rice biodiversity for impact, this project will involve extensive international collaboration and it has attracted considerable interest from partners and donors. Under the Rice SNP Consortium, funds have been raised from multiple sources, including public agencies (primarily the Japan Rice Breeding Project), research institutions, and the private sector, but additional resources are still needed. A one-time investment in modern, high-throughput phenotyping facilities is needed for the rapid evaluation of multiple traits. Ongoing funds will be needed to support large-scale trait evaluation for the phenotyping network. The genebank sequencing project will require major funding, especially from 2013 to 2015.
For bioinformatics support between 2010 and 2015, a minimum investment of $1,000,000 per year will be required. Fresh donor commitment to informatics for genetic resources is required, since designated funding is not currently in place. The genotype-phenotype data analysis is supported in part by the Japan Rice Breeding Project. Staff who are dedicated to software engineering are partially covered by the BMGF GCP “Molecular Breeding Platform” grant. However, to meet the challenges presented by rapidly expanding genotypic and genomic data sets, this activity will require additional funds above the baseline. Activities for comparative genomics analysis have begun in part through the C4 rice project. Extension of the concept to other research projects and engagement of ARI partners in a global network for comparative plant biology will require investment above the baseline. Long-term development of this resource may be considered for funding by the GIBS-GCP component of TA 3.