Genetic Architecture Of Grain Yield In Bread Wheat Based On Genome-wide Association Studies

Grain yield(GY)improvement is one of the most important and challenging objectives in wheat breeding due to the complex genetic architecture and low heritability.Marker-assisted selection(MAS)is considered to be a key technique to break through yield bottleneck of conventional breeding for further improvement of yield potential of wheat.Though many studies have focused on quantitative trait loci(QTL)mapping and genome-wide association study(GWAS)of yield and related traits,few loci or markers were used in the selection of wheat lines in breeding programs.This is mainly attributed to the low densities of molecular markers used in previous studies that resulted in large genetic distances between markers and QTL,as well as the complex genetic structure of GY.In the present study,a diverse panel of 166 wheat varieties from the Yellow and Huai River Valleys Wheat Zone(YHRVWZ)were grown at two locations for three consecutive years.The diverse panel was phenotyped for grain yield related traits and genotyped using the high-density wheat 90K and 660K SNP arrays.Two GWAS methods,i.e.SNP-GWAS and Haplotype-GWAS,were conducted based on the high-density physical map constructed with polymorphic SNP markers.The aims of the present study were to identify stable GY and related traits loci for MAS in breeding programs,and provide guidance for pyramiding favorable alleles,exploring the improvement directions of yield related traits and dissecting traits for rapid identification of high-yielding varieties.The main results were as follows:1.The diverse panel showed significant phenotypic variations for GY and related traits,and despite lower heritability(h~2)of GY,the h~2 of other traits were relatively high.The high-yielding varieties presented higher kernel number per spike(KNS),thousand-kernel weight(TKW),kernel width(KW),spike dry weight(SDW)and flag leaf width(FLW),and lower plant height(PH),uppermost internode length(UIL)and flag leaf length(FLL).2.In total,326,570 SNP markers from the 90K and 660K SNP arrays were used to construct a consensus physical map.These markers spanned a physical distance of 14,064.8 Mb with an average density of 0.043Mb per marker.The diverse panel consisted of three subgroups,and the average genetic diversity and polymorphism information content(PIC)were 0.34 and 0.27,respectively.The linkage disequilibrium(LD)decay distance was about 8 Mb for the whole genome,and 6,4 and 11 Mb for A,B and D genome,respectively.Genome-wide haplotype blocks were constructed based on 4 gametes,which generated 31,748blocks and 116,555 haplotypes.3.Three hundred and forty and 354 loci were identified on all 21 chromosomes by SNP-GWAS and Haplotype-GWAS,respectively,of which 150 loci were common between both methods.Among the common loci,two were potential functional genes underpin kernel weight and PH;10 were at the similar locations with the QTL identified in three RIL populations in our previous study,and 95 were likely to be new loci.The distribution of favorable alleles indicated that TKW,PH and flag leaf related traits were improved significantly in YHRVWZ,and the effects of favorable alleles on phenotypic values showed that the loci could reasonably explain the phenotypic variances among varieties.4.Seventeen pleiotropic loci were detected on 10 chromosomes,among which the interval 714.4–725.8 Mb on chromosome 3A was significantly associated with GY,KNS,KW,SDW,PH,UIL and FLL.GY shared five and six loci with TKW and PH,respectively,indicating that GY is largely affected by these two traits.More than half loci for kernel weight and shape related traits were co-localized with those for other traits,which indicated that the variation of kernel weight is easily caused by the altered agronomic traits.Moreover,20 PH and UIL loci were co-localized with the loci for other traits,showing considerable influence on other traits that investigated in the present study.5.Due to their simplicity,UIL and FLL can be used to select high-yielding varieties,whereas FLW is convenient to reflect the coordination degree of SN(spike number per unit area)and KNS.Several loci with positive allelic effects on different traits were identified on several chromosomes.For example,the SNP markers AX?110418502 on chromosome 1A,AX?94564150 and AX?109820171 on chromosome 1B,AX?111037158 on chromosome 2A,and AX?108921249 on chromosome 5B conditioned GY and PH.Similarly,the SNP markers AX?111634754 on chromosome 2B,IWB17930 on chromosome 3D,and AX?110523824 on chromosome 5A are significantly associated with GY and TKW.In general,the loci for GY and related traits and tightly linked molecular markers identified in the present study can be used for pyramiding favorable alleles to develop high-yield wheat cultivars.