Discovery of QTL Alleles for Grain Shape in the Japan-MAGIC Rice Population Using Haplotype Information

• Main Authors: Daisuke Ogawa, Yasunori Nonoue, Hiroshi Tsunematsu, Noriko Kanno, Toshio Yamamoto, Jun-ichi Yonemaru
• Format: Article
• Language: English
• Published: Oxford University Press 2018-11-01
• Series: G3: Genes, Genomes, Genetics
• Subjects: GWAS; haplotype; QTL; SNP; Multiparent Advanced Generation Inter-Cross (MAGIC); multiparental populations; MPP
• Online Access: http://g3journal.org/lookup/doi/10.1534/g3.118.200558

A majority of traits are determined by multiple quantitative trait loci (QTL) that can have pleiotropic effects. A multi-parent advanced generation inter-cross (MAGIC) population is well suited for genetically analyzing the effects of multiple QTL on traits of interest because it contains a higher number of QTL alleles than a biparental population. We previously produced the JAPAN-MAGIC (JAM) population, derived from eight rice (Oryza sativa L.) cultivars with high yield and biomass in Japan, and developed the method of genome-wide association study (GWAS) using haplotype information on the JAM lines. This method was effective for identifying major genes such as Waxy for eating quality and Sd1 for culm length. Here, we show that haplotype-based GWAS is also effective for the evaluation of multiple QTL with small effects on rice grain shape in the JAM lines. Although both the haplotype- and SNP-based GWAS identified multiple QTL for grain length and width, the sum of the estimated trait values of each allele for the QTL detected by haplotype-based GWAS had higher correlation with observed values than those detected by SNP-based GWAS, indicating high-accuracy QTL detection in the haplotype-based GWAS. Furthermore, the study revealed pleiotropic effects of some QTL regions in regulation of grain shape, suggesting that the haplotype-based GWAS using the JAM lines is an effective means to evaluate the main and side effects of haplotypes at each QTL. Information on the pleiotropic effects of haplotypes on various traits will be useful for designing ideal lines in a breeding program.