Next Generation Sequencing to Discover the Genome Diversity of Rice Germplasm.

Rice (Oryza sativa L.) is the staple food for more than half of the world population. From thousands of years of rice cultivation, China has accumulated an enormously diverse collection of landraces. This is an invaluable genetic resource for breeding elite varieties for sustainable agriculture. With high-quality reference genome sequences, rice is similar to the model plant, Arabidopsis, by having the self-fertilization system that allows simplified haplotype identification and repeated phenotyping for genome-wide association study (GWAS). There is strong evidence that GWAS of natural population represent a powerful approach to dissecting genes involved in important agronomic complex traits.

To further understand intra-specific variation and facilitate genetic improvement of rice, a comparative genomics approach will be necessary to make a more integrated and detailed map that collects all kinds of genetic variations. We sequenced 517 rice cultivated landraces and the identified ~3.6 million single nucleotide polymorphisms. An imputation method developed for low-coverage genome sequencing data was used to construct a high-density haplotype map of rice genome. Genome-wide association studies of comprehensively phenotyped and genotyped landraces identified eighty independent association signals for seven traits at P < 10^-7. Six traits were tied by signals to the previously identified genes. This study demonstrated an integrated approach of genotyping and association mapping, and provided a powerful resource to the community for extensively analyzing relationship between phenotypic and genetic variations in rice.