Linkage Analysis Of SSRs In F₂ Populations Of Rice(Oryza Sativa L) And Studies On Mapping Models Of Molecular Markers

Simple sequence repeats(SSRs, or microsatellite)is a codominant marker with sufficient polymorphisms. 312 SSRs were mapped on rice chromosomes with DH populations whose parent were japonica and indica. It is evident that F2 population is easer to be obtained than DH population. The number of markers with skewed segregation will be reduced by using intraspecific stains as parent. But their polymorphisms will be decreased at the same time. In order to construct SSRs linkage map with F2 population and to evaluate their distorted segregation, two F2 populations called cw2 and cw3 with same female parent 11 S, and Ux38, Ux42 as male parent respectively were produced. Their parent are all japonica rice. According to SSRs markers on rice chromosomes, 89 selected SSRs primers were investigated comparatively between male and female parent. Only 32 had polymorphisms between them. The segregations of the two F2 populations attributed differently: that of cw2 was distorted seriously and not suitable for map construction, but cw3 showed normal segregation and was fit for it. A 51.7cM long linkage group involving 5 SSRs nearby Wx gene was constructed and compared with the maps based on DH population of Temnykh抯 and Shen抯. Three point slefcross method is adapt to analysis linkage data of dominant markers. For the sake of the comparison to classical maximum likelihood(ML) method, detail steps of the two algorithms were listed for the estimations of genetic distances between genes. The results from calculations of dominate marker data came from rice RFLP data with three-point selfcross and two-point selfcross ML method exhibited that the linkage maps obtained were similar in marker order and distances with the map computed by software MAPMAKER /version 3.0 based on codominant data. But the distances between markers had the trend of being amplified when three-point selfcross method was used if there were evident distorted segregation in dominate marker data. Furthermore, the possibility of why segregation distortion can affect three-point sefcross method so much was discussed from the point view of the veracities of the two methods. At the same time, we found that using two-point selfcross ML method two times to estimate recombination fractions based on the dominate data changed from codominate data could simplify the process of the estimations. Segregation distortion of genetic markers is a common phenomenon. Larger bias occurs VII when skewed data treated with classical method. As to the equation systems considering zygotic and gametic selector, the estimation of recombination fraction will be more veracious. The original data used were 5 in 12 RFLPs of a mouse F2 population including 333 individuals presented in the software MAPMAKER. Classical and successive ~ tests were used to check segregation distortions of the codominant markers. And the result exhibited that successive ~ tests was more efficient. After the determinations of selection types, the suitable ML systems with parameter(gamete selected or zygote selected) were choosed to estimate recombination fractions between two markers with Newton-Raphson iteration method. The calculation also was compared with classical ML estimation. The results suggested that it is better to prefer the estimation considering segregation distortion if there were two linked markers with segregation distortions when constructing a molecular marker map, especially both of them were serious skewed. In the...