Genetic Dissection Of Rice Yield And Its Related Characters Utilizing Selective Introgression Lines

Rice is a very important crop globally, epecially in china, for it has very big planting arae and long-time history. Yielding more rice is always the seeking aim of rice production in china, even all through the world. The two "green revolution" of rice from 1950s, respectively labelled as rice dwarft and rice heterosis, markly promoted rice production per unit, resulting in much increase of gross output of rice, and consequently effectively relieving the severity of hunger. More and more demand of rice, caused by multiplying the earth and economic growth, requires continuous increase of rice production. However, the increase rice yield per unit much slows down from 1990s, causing rice prodution in state of plateform, much out of the way of rice demand, which presents a big challenge of rice prodution. Increasing rice prodution per unit is almost the single access to the settlement of the challenge, for the reason of iota room of rice planting expansion. Rice yield and its relating characters are complicated charaters, involving large number of genes and genetic mechanisms. Therefore, genetic dissection of Rice yield and its relating characters, gene excavation, and genetic machanism studying, all of which are aimed for rice breeding for better variety, are present ways of promoting rice production.To acqiure those,6 yield seletive backross population were constructed, separetely using Shuhui 527, distant blood-related to Minghui 63, Minghui 86, close blood-related to Minghui 63, as recurrent parents, using 3 accessories of ZD057, Fuhui 838 and Teqing as donor,2 times of self-cross after 2 times of back-cross, which followed the selecting better yielding plants. We studied the rice yield and its related characters, i.e. plant height(PH), parnicle length(PL), parnicle number per plant(PN), totle number of spikelets per parnicle(TNSP), number of filled grains per parnicle(NFGP), filling rate(FR), thousand-grain weight(TGW) and grain yield per plant(GYPP), explored the relationship among the characters, located the QTL by utilizing SSR mark though ANOVA, detected the epistasis effects and discussed the genetic network by applying the principle of linkage disequivilium.1. Correlation anlysis, path analysis, principal component analysis and stepwise regression analysis of rice yield and its related characters show: correlations of GRPP and the other characters is positive, while those among yield components are negetive; direct path coefficients of yield components to GYPP are positive to the significance level of 0.001, which is diminished by the negative indirect path coefficients, so that the co-promotion of the rice yield components is the only way of increase rice yield through breeding; through principle component analysis, mostly, the slopes of SNPP or NFGP are highest in the first principle compornts; TGW, PN and TNSP or NFGP contribute much to GYPP according to stepwise regression analysis.2. QTL location of rice yield and its related characters by utilizing SSR mark though one-way ANOVA shows:we got many QTL of TNSP and TGW and few of PN and GYPP; the location of QTL are strongly dependent on genetic background and planting enviroment; the QTL, which was located in the same way in the two trials of the same population, is worthy of being applied in rice breeding. In the same population, the QTLs of close related characters are located at the same loci or close-linked loci, for example, which is the genetic base of high correlation coefficient between the characters; from the panorama of this QTL location, the QTLs of rice yield and its related characters are cluster-distribing, the close-linked marks of RM282, RM2334 and RM16 and the other close-marks of RM3 and RM5957 are examples, the counterpart chromosome are important in rice breeding.35 pairs of epistasas interaction of rice yield and its related characters in total at the level of p<0.00001 were detected though two-way ANOVA. The loci partipaciting in epistasas are distributed in cluster. QTLs of times of participation in epstasas are important in rice yield development, especially the mark RM16, eight times of QTL location and 5 times of epstasis, becaming the "hot" locas in the present study.3. The Linkage Disequilibrium was conducted on the 6 populations in 2 enviroments, by selecting lines of yielding more than the recurrent parents, which follows the connection of 2 marks with LD in low p-value, and the genetic network generally constructing. The genetic network of rice yield varies, depending on genetic background and eviroment, showing the difference in genetics and motablism. QTLs actively participating in the genetic network take very important roles in rice yield development. Geneally, genetic networks derived from Hefei planting season are more complex those from Hainan season, possibly for the reason of the line selection of the populations mainly according to the yield of the lines planted in Hefei. The genetic networks of Shu-Fu and Ming-ZDZ are very complex, fitting for the study of rice yield development, so that further dessetion on the yield characters of the 2 populations can get more complex mechanism of yield development.