QTL Analysis For Controlling Yield Traits In Dongxiang Wild Rice(Oryza Rufipogon Griff.)

In this paper, We used Dongxiang wild rice （O. rufipogon Griff., hereafter Dwr）as the donor parent, and the cultivar Xieqingzao B （O. sativa sp. indica Kato,hereafter Xqz B） as the recurrent parent to construct a BILs （backcross inbred lines）population BC₁F₁₀including234lines. And these materials were employed toinvestigate and analyze the yield characteristics, including total Spikelets per panicle（SP）, Spikelets per panicle （FSP）, Effective tillers number per plant （ETNP）,Thousand grain weight （TGW）, Seed setting rate （SSR）, Panicle length （PL）, Panicledensity （PD）, Grain weight per plant （GWP）. Meanwhile, by using constructedmolecular linkage map, we mapped the related QTL locus from Dongxiang wild rice.The main results were summarized as follows:1. Genetic analysis of yield characteristicsThe BC₁F₁₀population was employed to evaluate the eight major yieldcharacteristics. Based on the comprehensive analysis, we found that the phenotypes ofthese traits exhibited continuous distributional, which suggested that the inheritance ofthese traits were controlled by qualitative-quantitative traits.Linkage correlation relationships were still existed among these yield traits. Inthis study, the relationships between the GWP and other eight yield characteristicswere analyzed, and all traits showed distinctly relationship （0.26⁰.90）. Therelationship coefficient between various traits and GWP were0.90,0.77,0.63,0.44,0.42,0.34,0.27and0.26, respectively. The degree of relationship was FSP> SP> PD> SSR> TGW> ETNP> PL.Among that, the correlationship between filled grains per panicle and spikeletsper panicle, grain weight per plant and filled grains per panicle, panicle density andspikelets per panicle showed distinctly relationship and all more than0.90. Therefore,we suggested that, in the process of rice breeding, the filled grains per panicle andspikelets per panicle real should be paid more attention, and the combined with seedsetting rate, thousand grain weight, effective tillers number per plant and paniclelength as the major reference indices for selecting.2. QTL mapping of yield characteristicsBy using130pairs of SSR markers, the genetic linkage map of the BC₁F₁₀population was constructed. Then, we used the established genetic linkage map to analyze the yield characteristics QTLs. A total of26QTLs were detected, including2QTLs for spikelets per panicle,3for filled grains per panicle,1for effective tillersnumber per plant,6for seed setting rate,3for panicle length,7for panicle density and3for grain weight per plant. The28QTLs were distributed at all12rice chromosomesexcept chromosomes5,8,9. A QTL for filled grains per panicle in the12chromosome with enhancing alleles was from the Dwr, which was similar to theresults from LiDeJun. A QTL for spikelets per panicle located in the1chromosomeRM10176-RM243was Qnsp1-1, which was similar to the results from LiDeJun andZhuangJieYun. Among these QTLs,13QTLs with enhancing alleles were from theDwr, including6QTLs for thousand grain weight. Meanwhile,2QTLs for spikeletsper panicle with enhancing alleles were from Xqz B.The results from this study laid a solid foundation for further fine mapping andcloning the related genes of yield characteristics in Dwr, and its utilization in ricebreeding with marker-assisted selection.