QTL Analysis for Yield in a Japonica Rice (Oryza Sativa L) RIL Population
|School||Nanjing Agricultural College|
|Keywords||japonica rice yield conditional QTL flag leaf lamina joint|
The human population will explode to 9 billion at 2050, thus increasing the yield of crops to feed the more and more people is the major task for breeders. Considering that 50% of the population takes rice as their stable food, it is very important to breeding high productive rice cultivars. Between the two major cultivated subspecies indica and japonica, japonica rice yields 14% higher than indica rice does, and is grown annually 8 million hm2 in china. Thus finding the genetic factors affecting the yield of japonica will lead us to effectively improve the productivity of rice. Obviously, yield is one of the most complex traits which is releated to many traits and the investigating of yield, in fact, contained the influence of other traits. In order to eliminate the influence of related traits, in the first part of this thesis, we used a conditional and an unconditional QTL method to analyse the yield and yield related traits of a intra-subspecies RIL population derived from a cross of Xiushui 79 XC Bao in 3 different environments which can help us to dissect the relations between yield and yield related trais. Results obtained were as follows.1. By unconditional method, totally 30 main-effect QTLs (M-QTLs) and 22 pairs of epistatic QTLs (E-QTLs) were detected. M-QTLs were located on 11 chromosomes except chromosome 12, while E-QTLs spread in all 12 chromosomes.6 M-QTLs and 8 pairs of E-QTLs were detected for plant height (PH) which explained 72.5% of the phenotype variance.6 M-QTLs and 9 pairs of E-QTLs were detected for whole growth period (WGP) which explained 41.1% of the phenotype variance.7 M-QTLs and 2 pairs of E-QTLs were detected for the number of panicles per plant (PN) which explained 30.4% of the phenotype variance.7 M-QTLs and 4 pairs of E-QTLs were detected for the number of grains per panicle which explained 33.5% of the phenotype variance.4 M-QTLs and 1 pair of E-QTLs were detected for hundred kernels weight (HKW) which explained 35.65% of the phenotype variance.4 M-QTLs and 2 pairs of E-QTLs were detected for bimass yield per plant (BY) which explained 20.63% of the phenotype variance. Only 3 M-QTLs were detected for grain yield per plant (GY) which explained 12.25% of the phenotype variance.2. We totally found 12 pleiotropic regions which affected more than one trait. By comparing the results with previous researches, we found the marker interval RM7097-RM448 located on the long arm of chromosome 3, RM3288-RM303 located on the long arm of chromosome 4, RM8239-RM5314 located on the long arm of chromosome 6, RM22899-RM22957 located on the long arm of chromosome 8 and RM6570-RM5652 located on the long arm of chromoseome 9 were newly detected regions.3.9 M-QTLs and 3 pairs E-QTL were found by conditional method including the 3 M-QTLs detected by unconditional method. The interval RM5652-RM410 located on the long arm of chromosome 9, which could be found by unconditional method, had similar additive effect and variance explained when GY was conditioned on WGP. And it had decreased additive effect and variance explained when GY was conditioned on PH, HKW, PN, but it had increased additive effect and variance explained when GY was conditioned on GN. Thus the allele from C Bao of this region could increase GY through increasing PH, HKW and PN.4. The marker interval RM7097-RM448 for GY located on the long arm of chromosome 3 deteced by conditional method also had additive effect for PH and WGP, but the direction was different that the allele from Xiushui 79 can increase GY but decrease WGP and PH. The marker interval RM162-RM5753 located on the long arm of chromosome 3 also had opposite effect for GY and WGP that the allele form C Bao can increase GY but decrease WGP. Condidering that the PH and WGP should be controlled in a appropriate scale, these QTLs is practical in rice breeding.