Genomic regions involved in yield potential detected by genome-wide association analysis in Japanese high-yielding rice cultivars
- Equal contributors
1 National Institute of Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan
2 NARO Institute of Crop Science, 2-1-18 Kannondai, Tsukuba, Ibaraki 305-8518, Japan
3 NARO Institute of Vegetable and Tea Science, 360 Kusawa, Ano, Tsu, Mie 514-2392, Japan
4 NARO Tohoku Agricultural Research Center, 3 Yotsusya, Daisen, Akita 014-0102, Japan
5 NARO Hokuriku Agricultural Research Center, 1-2-1 Inada, Jyoetsu, Niigata 943-0193, Japan
BMC Genomics 2014, 15:346 doi:10.1186/1471-2164-15-346Published: 8 May 2014
High-yielding cultivars of rice (Oryza sativa L.) have been developed in Japan from crosses between overseas indica and domestic japonica cultivars. Recently, next-generation sequencing technology and high-throughput genotyping systems have shown many single-nucleotide polymorphisms (SNPs) that are proving useful for detailed analysis of genome composition. These SNPs can be used in genome-wide association studies to detect candidate genome regions associated with economically important traits. In this study, we used a custom SNP set to identify introgressed chromosomal regions in a set of high-yielding Japanese rice cultivars, and we performed an association study to identify genome regions associated with yield.
An informative set of 1152 SNPs was established by screening 14 high-yielding or primary ancestral cultivars for 5760 validated SNPs. Analysis of the population structure of high-yielding cultivars showed three genome types: japonica-type, indica-type and a mixture of the two. SNP allele frequencies showed several regions derived predominantly from one of the two parental genome types. Distinct regions skewed for the presence of parental alleles were observed on chromosomes 1, 2, 7, 8, 11 and 12 (indica) and on chromosomes 1, 2 and 6 (japonica). A possible relationship between these introgressed regions and six yield traits (blast susceptibility, heading date, length of unhusked seeds, number of panicles, surface area of unhusked seeds and 1000-grain weight) was detected in eight genome regions dominated by alleles of one parental origin. Two of these regions were near Ghd7, a heading date locus, and Pi-ta, a blast resistance locus. The allele types (i.e., japonica or indica) of significant SNPs coincided with those previously reported for candidate genes Ghd7 and Pi-ta.
Introgression breeding is an established strategy for the accumulation of QTLs and genes controlling high yield. Our custom SNP set is an effective tool for the identification of introgressed genome regions from a particular genetic background. This study demonstrates that changes in genome structure occurred during artificial selection for high yield, and provides information on several genomic regions associated with yield performance.