Plant & Animal Genomes XVIII Conference

January 9-13, 2010
Town & Country Convention Center
San Diego, CA

Population Structure And Association Mapping In Watermelon Heirloom Collections

Padmavathi Nimmakayala¹ , Yan R Tomason^{1, 2} , Viktoria K Sokolova³ , Gopinath Vajja¹ , Sathish K Ponniah¹ , Umesh K Reddy¹

¹  Department of Biology, Gus R. Douglass Institute, West Virginia State University, Institute, WV 25112-1000, USA

²  Department of Plant Breeding, Dnepropetrovsk State Agrarian University, Voroshilov 25, Dnepropetrovsk 49600, Ukraine

³  Institute of South Vegetables and Melon Crops, UAAS, Gola Pristan 75600, Ukraine

Thirty five watermelon heirlooms from the USA, Ukraine and Russia were evaluated for three seasons for their growth and fruit traits. The genotypic data generated by SSRs and several hundreds of AFLPs was utilized to resolve population structure using STRUCTURE program. This analysis was conducted assuming two subpopulations (K=2) to five subpopulations (K=5) using the SSR data and assuming two subpopulations (K=2) to eight subpopulations (K=8) using the AFLP data. The results indicated presence of subpopulation structure within the heirlooms that ranged from 3 to 6 clusters. In all the three runs with the K=3 assumptions with the SSR and AFLP data, there were 3 clusters: one with mixture of heirlooms from all the countries and the second with the US and the third with the Russian and Ukraine heirlooms showing a sub structure of 18,8 and 9 heirlooms separately. In all the three runs of K=6 assumptions using the AFLP data, there were 6 clusters. The clustering based on the assumptions K=8 was not any additionally informative but conferred to the same pattern of clustering. The clustering results (six subgroups) were used as covariates in the association test (MLM procedure with TASSEL software). For the traits with high heritability values (> 0.5), i.e. rind pressure and soluble solids 4 to 8 markers were identified showing significant associations in up to 2 out of 3 seasons, with R² values ranging from 5 to 10%. As to fruit yield and yield components, the majority of the markers could be identified with R² values lower than 5%. Detailed results, including co-location of markers significant for different traits will be presented,