Open access resources for genome-wide association mapping in rice.
McCouch SR, Wright MH, Tung CW, Maron LG, McNally KL, Fitzgerald M, Singh N, DeClerck G, Agosto-Perez F, Korniliev P, Greenberg AJ, Naredo ME, Mercado SM, Harrington SE, Shi Y, Branchini DA, Kuser-Falcão PR, Leung H, Ebana K, Yano M, Eizenga G, McClung A, Mezey J. · 2016
Read Open access resources for genome-wide association mapping in rice. by McCouch SR, Wright MH, Tung CW, Maron LG, McNally KL, Fitzgerald M, Singh N, DeClerck G, Agosto-Perez F, Korniliev P, Greenberg AJ, Naredo ME, Mercado SM, Harrington SE, Shi Y, Branchini DA, Kuser-Falcão PR, Leung H, Ebana K, Yano M, Eizenga G, McClung A, Mezey J. (2016) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4742900.txt.
Introduction
McCouch and 25 colleagues launch an open-access platform for genome-wide association mapping in rice, the staple crop that feeds billions. Breeders need fresh genetic variation to raise yields and resilience. The platform pairs a permanent, diverse germplasm collection with a high-density marker dataset and documented analysis strategies.
The team demonstrates the system on grain length. It detects major and minor effect loci, including alleles specific to rice subpopulations, at fine resolution. Because the germplasm never changes, results from different studies can be compared and stacked. The authors offer the resource as shared infrastructure for gene discovery and rice improvement worldwide.
Catalog record
| CREATOR | McCouch SR, Wright MH, Tung CW, Maron LG, McNally KL, Fitzgerald M, Singh N, DeClerck G, Agosto-Perez F, Korniliev P, Greenberg AJ, Naredo ME, Mercado SM, Harrington SE, Shi Y, Branchini DA, Kuser-Falcão PR, Leung H, Ebana K, Yano M, Eizenga G, McClung A, Mezey J. |
|---|---|
| YEAR | 2016 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 70 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
By Susan R McCouch, Mark H Wright, Chih-Wei Tung, Lyza G Maron, Kenneth L McNally, Melissa Fitzgerald, Namrata Singh, Genevieve DeClerck, Francisco Agosto-Perez, Pavel Korniliev, Anthony J Greenberg, Ma Elizabeth B Naredo Published in: Nature Communications DOI: 10.1038/ncomms10532 ABSTRACT Abstract Increasing food production is essential to meet the demands of a growing human population, with its rising income levels and nutritional expectations. To address the demand, plant breeders seek new sources of genetic variation to enhance the productivity, sustainability and resilience of crop varieties. Here we launch a high-resolution, open-access research platform to facilitate genome-wide association mapping in rice, a staple food crop. The platform provides an immortal collection of diverse germplasm, a high-density single-nucleotide polymorphism data set tailored for gene discovery, well-documented analytical strategies, and a suite of bioinformatics resources to facilitate biological interpretation. Using grain length, we demonstrate the power and resolution of our new high-density rice array, the accompanying genotypic data set, and an expanded diversity panel for detecting major and minor effect QTLs and subpopulation-specific alleles, with immediate implications for rice improvement. Rice (Oryza sativa) is a staple food that feeds 3 billion people1,2. It has the largest, publicly available, single-species germplasm collection in the world3, and a rich ‘omics' toolkit. Rice was the first crop plant to be fully sequenced4, has over 3,000 re-sequenced varieties5,6,7,8, de novo assemblies within three subpopulations9,10,11 and BAC-based sequence assemblies for 12 related Oryza species12. With these resources in hand, the challenge is to utilize this information to understand and predict how genotypic variation gives rise to the abundance of phenotypic variation that we see in nature and in agriculture. Genome-wide association studies (GWAS) offer a powerful strategy for understanding the genetic basis of complex traits. This approach has been particularly productive for rice, where major associations have been detected for important traits, including grain size, grain weight, flowering time, disease resistance and abiotic stress tolerance, among others13,14,15,16. Hundreds of genes governing traits of interest have been cloned and new, powerful tools are currently being deployed to achieve targeted genome editing17,18,19. In applied breeding programs, major-effect loci discovered via quantitative trait loci (QTL) mapping and GWAS have been successfully introgressed from wild and exotic genetic resources into elite, high-yielding varieties via marker-assisted selection20,21,22. Rice has been particularly amenable to GWAS largely because of its evolutionary and domestication history. The wild ancestor of Asian rice, O. rufipogon, is a partially (35%) outcrossing species complex broadly dispersed across tropical South, South-East and Eastern Asia. It consists of locally or regionally adapted subpopulations with both perennial and annual forms23. The domestication of O. sativa began ∼10,000–12,000 years ago, and human selection was accelerated by inbreeding which retained and fixed many favourable alleles of large effect. The unique combinations of locally adapted allele complexes in different subpopulations provide ideal characteristics for GWAS, including both universal and population-specific large effect alleles, extended linkage disequilibrium (LD) within some subpopulations that is advantageous for mapping with low marker coverage, rapid LD decay within other subpopulations offering improved mapping resolution and an abundance of well-partitioned genetic variation among subpopulations. These properties, in combination with a small, tractable genome (380 Mb), and the ability to produce immortal panels comprised of thousands of inbred, homozygous individuals, make it possible to evaluate the same traits in different environments simultaneously, and to evaluate a variety of different traits on the same germplasm over time. The ability to make controlled matings and to generate large populations each generation provides a powerful system for validating loci and for harnessing their value by introgressing favourable alleles into other genetic backgrounds. Here, we report the development of an association mapping panel consisting of 1,568 diverse inbred rice varieties and a genotyping data set generated using a high-density rice array (HDRA) comprised of 700,000 single-nucleotide polymorphisms (SNPs). The HDRA represents the most densely populated genotyping array available for any organism at this time, based on the number of SNPs/kb interrogated across the genome. We explore the value of these resources for GWAS using grain length as the phenotype, and use numerous analytical techniques to identify significant associations that uncover both major and minor QTLs. Further, we demonstrate distinct advantages of cross-population mapping where we gain power by directly accounting for large effect alleles from within the mapping model. We show how this may lead to the identification of population-specific haplotypes, which are of particular value to the plant breeder. The GWAS resources and data sets generated in this project are designed to facilitate the collection of phenotypic information on a wide range of traits and characteristics (see companion paper by Crowell et al.24) and to rapidly and efficiently convert that information into focused hypotheses about causal genes and QTL locations. They also provide a bridge between genomics and applied plant breeding. With increasing accuracy over time, data sets generated for different purposes and at differing levels of resolution can be leveraged against each other via shared haplotypes, making it feasible to undertake cross-population imputation of both genotype and phenotype and to predict the phenotypic outcomes from genotypic data. GWAS results can also be incorporated as fixed variables into genomic selection models to improve the accuracy of genomic estimated breeding values25,26,27. These resources are part of a larger rice diversity research platform being developed by the international rice research community. All the materials, data sets and tools reported here are available as open-source products. Together these resources provide the basis for further exploration of phenotype–genotype relationships in rice and for expediting variety development in this important food crop. RESULTS The high-density rice array The HDRA assays 700,000 SNPs (average call rate=93.5%), or ∼1 SNP every 0.54 kb across the rice genome. It was designed to efficiently capture most of the haplotype variation observed in a discovery panel consisting of 16M SNPs (generated by sequencing 125 rice genomes at ∼7 × genome coverage) and to maximize the inclusion of non-synonymous SNPs (see Supplementary Note 1). An estimated 45% of HDRA SNPs map within genes, covering all 39,045 unique, non-TE rice gene models (http://rice.plantbiology.msu.edu/), while 55% of SNPs map to intergenic regions (Supplementary Fig. 1). Non-synonymous SNPs, which represent ∼16% of all informative SNPs on the HDRA (Table 1), are found in 91% of unique, non-TE gene models. Of the genic SNPs, 55% are distributed within exons, 36% within introns, 3% within 5′-untranslated regions (UTRs) and 6% within 3′-UTRs. Of the intergenic SNPs, 50% are located in putative regulatory regions (PPRs) within 2 kb of a transcriptional start site. [Table: Summary of polymorphisms detected by the HDRA in the different rice subpopulations.] Genetic structure of the rice diversity panel Using the HDRA, we assayed SNP variation in a collection of 1,568 diverse accessions of O. sativa (Supplementary Data 1) and the resulting genotypic data set was analyzed for population structure using fastSTRUCTURE28. The optimum number of subpopulations was determined to be K=5. The five subpopulations were identified as aus, indica, tropical japonica, temperate japonica and aromatic (or Kashmiri japonica, Group V), with aus and indica comprising the INDICA varietal group, and tropical japonica, temperate japonica and aromatic comprising the JAPONICA varietal group15,16,29,30. (We use upper-case letters to indicate the two major varietal groups, INDICA and JAPONICA, and lower-case letters to