Correlation between molecular marker distance and hybrid performance in U.S. Southern long grain rice
Saghai Maroof, M. A., G. P. Yang, Q. Zhang and K. A. Gravois. 1997. Crop Science 37:145-150.
ABSTRACT Improving long-grain rice yields and milling yields continues to be a major objective of southern United States rice breeding programs. The objective of this study was to assess the relationship between molecular marker polymorphism of the parents and hybrid performance. A diallel cross was made by intermating eight lines commonly used in US southern long-grain rice breeding programs. The 28 F1 hybrids and the parents were evaluated for performance of three traits (rough rice yield, head rice percentage and head rice yield) in replicated field trials at two locations in Arkansas. Heterosis varied widely from one trait to another; very little heterosis was detected among hybrids for head rice percentage, whereas large and highly variable amounts of heterosis were observed in rough rice yield and head rice yield. The eight parental lines were surveyed for DNA polymorphisms with two classes of molecular markers, 82 RFLPs and 26 microsatellites distributed along all 12 chromosomes, which revealed high levels of polymorphism among the parents. A cluster analysis resolved these eight lines into two major groups. A majority of the markers detected statistically significant effects on one or more traits by an analysis of variance. F1 heterozygosity, calculated as percentage of difference between the parents in marker genotypes, is highly correlated with rough rice yield and head rice yield, and also significantly correlated with heterosis of these two traits. In comparison to the data published in previous studies, it is concluded that the level of correlations between marker distance and hybrid performance is dependent upon the materials used in the studies.
Key words Oryza sativa, RFLP, SSR, heterosis, diallel cross