Genetic structure of the Japanese soybean population

The genetic structure of the Japanese soybean population was evaluated based on the variation at 16 isozyme and seed protein loci for 781 landraces and pure line selections. A relatively high diversity was found in the Japanese population, compared with 158 Korean accessions and 94 Chinese accessions that were tested for comparison. The average gene diversity in the Japanese population was 0.248, being comparable to the estimate obtained for the Chinese population (0.249) but higher than that for the Korean population (0.209). There were a number of alleles that differentiated the Japanese population from either of the Korean or Chinese populations or both. The presence of the alleles which are characteristic of the Japanese population suggests that the Japanese population is not solely a part of the population of China, which is considered as the place of origin of soybean. Only 11% of the total variation was attributable to the differences among seven regional groups, which occurred mainly between Hokkaido or Okinawa and the other regions. A test of two-locus associations revealed that the variation was itself highly structured despite the high level of variation at each of the polymorphic loci. In addition, Aco1, Aco4, Dia1, Enp and Idh1 varied with maturity or seed size. The Japanese soybean population may consist of different cultivar groups that had been established independently as the result of adaptation to different environmental conditions and/or diversification of food culture. The origins of the cultivar groups are discussed.