NARO performs large-scale soybean genome analysis: Different cultivar characteristics between Japan and the US

The National Agriculture and Food Research Organization (NARO) has sequenced the whole genomes of 11 soybean cultivars, including representative Japanese soybeans, and released them as reference genome data. By developing and applying a pangenome analysis method that analyzes genomic structural variations (SVs) at the gene level along with this genomic information, they conducted a large-scale analysis of 462 soybean accessions from Japan and around the world. As a result, they revealed that Japanese soybeans have genomic characteristics that differ from global cultivars, particularly US cultivars. They also identified genomic regions and genes related to the seed shape characteristics of Japanese cultivars—a large and plump round shape. The findings were published in Nature Genetics.

Japanese cultivars have good seed appearance and high protein content and are primarily used as food. On the other hand, US cultivars have yields approximately twice as high as Japanese varieties, high oil content, and strong resistance to stresses such as drought, and are mainly used as raw materials for oils, livestock feed, and chemicals.

Director Masao Ishimoto of the Institute of Crop Science, stated: "The domestic share of food soybeans in Japan is 24%, but we conducted comparative genomic analysis because we want to develop cultivars that combine high yield, stability, and high quality to aim for complete technical self-sufficiency."

First, they assembled 20 chromosomal genomic DNAs for each of the 11 cultivars and identified approximately 60,000 genes present in the genome. While SNPs and short insertion-deletion mutations can be easily analyzed with short reads, analysis of SVs where entire genes are deleted is difficult, so pangenome analysis was performed. Once a pangenome is constructed, comprehensive SV analysis of hundreds of lines can be performed from inexpensive short-read data.

For this study, the group developed a pangenome construction method (Asm2sv) for precise analysis of (SVs) per gene. They succeeded in identifying 387 SVs with genes that could not be found using conventional methods.

Analysis of 140 domestic cultivars and 322 foreign cultivars using this data revealed particularly large differences between Japanese and US cultivars. The researchers discovered that relatively traditional Japanese endemic cultivars such as Tanba and Dadacha-mame share common sequence types in five genomic regions.

Principal Researcher Ryoichi Yano at the Research Center for Advanced Analysis, said: "By further identifying genes related to traits, we want to connect this to smart breeding that selects innovative new cultivars with both high yield and high quality from various crossed varieties."

The gene function information, atlas, sequence information, and DNA mutations published in the paper are available in the genome database "Daizu-net."

Journal Information
Publication: Nature Genetics
Title: The genomic landscape of gene-level structural variations in Japanese and global soybean Glycine max cultivars
DOI: 10.1038/s41588-025-02113-5