Assessing the genetic diversity of rice originating from Bangladesh, Assam and West Bengal

Background: Genetic diversity among rice cultivars from Bangladesh and North East India was assessed using a custom 384-SNP microarray assay. A total of 511 cultivars were obtained from several sources, choosing landraces likely to be from the aus subpopulation and modern improved cultivars from Bangladesh. Cultivars from the OryzaSNP set and Rice Diversity Panel 1 (RDP1) were also included for reference. Results: The population analysis program STRUCTURE was used to infer putative population groups in the panel, revealing four groups: indica (76 cultivars), japonica (55) and two distinct groups within the aus subpopulation (aus-1 = 99, aus-2 = 151). Principal Component Analysis was used to confirm the four population groups identified by STRUCTURE. The analysis revealed cultivars that belonged to neither aus-1 nor aus-2 but which are clearly aus based on the combined probabilities of their membership of the two aus groups which have been termed aus-admix (96). Information obtained from the panel of 511 cultivars was used to assign rice groups to 74 additional landraces obtained from Assam and West Bengal. While both the aus-1 and aus-2 groups were represented approximately equally in India, aus-2 (which includes cultivar N 22) was more common in Bangladesh, but was not found at all in West Bengal. Conclusions: Examining the distribution of landrace names within theaus-1 and aus-2 groups suggests that aus-1 is associated with the term “boro”, a word used to describe a winter growing season in Bangladesh and Assam. The information described here has been used to select a population of 300 cultivars for Genome Wide Association studies of the aus rice subpopulation.