Submitted to: Crop Science Congress
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/26/2004
Publication Date: 7/26/2004
Citation: Lu, H., Redus, M.A., Coburn, J.R., Rutger, J.N., McCouch, S.R., Tai, T.H. 2004. Population structure and breeding patterns of 145 US rice cultivars based on SSR marker analysis. Crop Science. In Press.
Interpretive Summary: Using DNA markers covering the 12 chromosomes of rice, the genetic diversity of 145 US rice cultivars was examined. US rice cultivars are primarily of the subspecies japonica and fall into three groups: temperate, tropical medium grain, and tropical long grain. Some cultivars of the indica subspecies have contributed to the development of US varieties. Analysis of cultivars from the early 1900s to 2000 indicates that the population structure or make up of US rice was established prior to 1930 and remains more or less intact in the present day. Over half of the cultivars examined were developed from within group crosses. This indicates that there is substantial genetic variability within each group. Two different methods for assessing genetic diversity were compared and discussed. Similar results were obtained with each method. The data presented here in conjunction with phenotypic information from these cultivars provides a foundation for future application of association mapping to identify chromosomal regions controlling important agricultural traits.
Technical Abstract: This study was undertaken to invesitgate the population structure of US rice. A total of 115 US rice cultivars and 30 ancestral accessions introduced from Asia were genotyped using 169 simple sequence repeat (SSR) markers that are well distributed throughout the rice genome. SSR-based clustering analysis identified three groups of US rice cultivars that were recognizable as temperate japonicas with short to medium grains, tropical japonicas with medium grains, and tropical japonicas with long grains. Indica cultivars were represented among ancestral accessions but always clustered independently, having been used for cultivar improvement but never directly in US rice production. When US rices were divided into four time periods (1900-1929; 1930-1959; 1960-1979; and 1980-2000) according to the date of their first release or introduction, and clustering was performed within each time window, the same three groups were identified. This suggests that the population structure in US rice was established prior to 1930 and remains essentially intact today, despite a large amount of controlled crossing and artificial selection as a part of the breeding process. Fifty seven percent of US rice cultivars were developed from intra-group crosses, indicating the availability of substantial genetic variability within each group. Information about population structure and associated phenotypic characteristics recognized by geneticists and breeders paves the way for coordinated association mapping studies in the future. Genetic distance-based and model-based clustering methods were compared and discussed in this study.