Mapping Of QTLS For Grain Traits And Rapid Visco-analyze Profile In Rice

Based on a recombinant inbred (RI) population consisting of 131 lines drived from a cross between two indica rice vareties (H359and Acc8558) and a corresponding genetic map comprising 147 RFLP markers and 78 AFLP markers, QTLs controlling grain length (GL), grain breadth (GB), grain weight (GW) (rice apprearance qualtity) and rice rapaid visco-analyse profikle (RVA) (rice cooking quality) were analyzed. The results are summaried as follows:(1) Fiffteen QTLs for GL were detected on 8 chromosomes, which could expain 71.91% of the phenotypic variance. Among the QTLs, four on chromosomes 3, 4, 7 and 10 had the largest additive effects and could explain 7.47%, 8.02%, 7.43% and 8.08% of the phenotypic variance, respectively.(2) Seventeen QTLs for GB were detected on 9 chromosomes, which could expain 76.2% of the phenotypic variance. A QTL on chromosome 5 had the largest additive effect and could explain 26.65% of the total phenotypic variance.(3) Sixteen QTLs for GW were detected on 8 chromosomes, which could explain 81.4% of the phenotypic variance. Among the QTLs, two on chromosomes 5 and 8 had the largest additive effects and could explain 13.31% and 9.51% of the total phenotypic variance, respectively.(4) Highly significant correlations were found between the three characteric parameters of RUA profile, i.e. peak viscosity (PKV), hot paste viscosity (HPK) and cool paste viscosity (CPV), suggesting that the three traits have common genetic basis. Therefore, the three traits were jointly analyzed for QTL mapping. Twenty-four QTLs were detected, which could explain 56.27%, 72.24% and 67.51% of the total phenotypic variances of PKV, HPV and CPV, respectively. QTL effects and proportions of explained phenolypic variance were also highly correlated between the three traits. This is consistent with the phenotypic correlations between the three traits.(5) Although the correlation between GL and GB was statistically significant, the correlation coefficient was very small (r = 0.179*). In addition, almost no common QTLs were found between the two traits. These results suggest that GL and GB have relatively independent genetic basis. The correlation coefficient between GL and GW(0.781) and the path coefficient from GL to GW (0.722) were greater than that between GB and GW (0.461) and that from GB to GW (0.330). QTL mapping also showed that GW shared more common QTLs with GL than with GB. These results suggest that GL contributes more to GW.(6) There were no marked correlation and very few similar QTLs between the appearance quality (including GL, GB and GW) and the cooking quality (including PKV, HPV and CPV), indicating that the two types of grain quality traits have relatively independent genetic basis.