Genome-wide Association Study For Agronomic Traits And Grain Traits In Sichuan Wheat

Wheat(Triticum aestivum L.)is one of the main grain crops in China.Breeding new wheat varieties with high yield,disease resistance and high quality is an important means to ensure the safety of wheat production.In the past 20 years(1997-2016),165 wheat varieties have been examined and approved in Sichuan wheat region,which is one of the six major wheat producing areas in China.These varieties have made outstanding contributions to wheat production in Sichuan.In this study,agronomic and grain traits of the Sichuan wheat germplasm population which composed of 156 wheat varieties approved in Sichuan Province and 77 Sichuan wheat landraces were evaluated in multiple environments.A scanning with genome-wide wheat 55 K SNP(single nucleotide polymorphism)microarray was carried out to parse the genetic diversity,population structure and the degree of linkage disequilibrium of Sichuan wheat germplasm population,reveal the genetic basis in molecular level and provide theoretical basis for breeding utilization.At the same time,genome-wide association analysis was used to identify loci or QTL controlling agronomic traits or grain traits of wheat.This assessment provided a basis for efficient utilization of genes related to important breeding target traits.The main research results are as follows:1.In three environments in 2019(Wenjiang,Chongzhou and Ya’an),the agronomic traits and grain traits were evaluated in 233 Sichuan wheat germplasms.The results showed rich phenotypic diversity of the agronomic traits and grain traits of Sichuan wheat germplasm.Significant differences were detected between landraces and varieties in plant height,number of effective tillers,awn length,grain number per spike,thousand kernel weight and linear length and width of grains,except the number of fertile spikelets.The average values of number of effective tillers,plant height,total spikelet number,infertile spikelet number,spikelet density,grain number per spikelet and grain number per spike in Sichuan wheat landraces were all higher than that of cultivated varieties.While the average values of spike length,thousand kernel weight and linear length and width of grains in Sichuan wheat landraces were lower than that of cultivated varieties.Among these traits,landraces(7.39)> cultivated variety(5.72)in number of effective tillers,cultivated variety(47.27g)> landraces(37.43g)in thousand kernel weight.11 germplasms with big spike(grain weight per spike≥2.1g,spike length≥10cm,grain number per spike≥60,1000-grain weight≥45g)were obtained,all of which were cultivated varieties.19 germplasms with dense spikelet(panicle density≥2.2)were obtained,including 10 landraces and 9 cultivated varieties.19 germplasms with multi-tillers(number of effective tillers≥10)were obtained,of which 7 were cultivated varieties and 12 were landraces.15 cultivated varieties with high 1000-grain weight(1000-grain weight≥50g)were obtained.These excellent varieties can be used in wheat breeding.2.Sichuan wheat germplasm was scanned by genome-wide 55 K SNP markers.A total of 44,059 effective markers were obtained,of which the maximum number was on B genome with 16,770 markers,and the minimum number was on D genome with 11,263 markers.The distribution of the second homologous group was the most(6,849)and the fourth homologous group was the least(5,519).At the chromosomal level,the maximum number was on chromosome 2A with 2547 markers,and the minimum number was on chromosome 4D with 944 markers.The genome-wide gene diversity was 0.380,and the polymorphism information content index(PIC)was 0.301.The B genome showed the highest gene diversity and PIC(0.396 and 0.312),and the D genome PIC the lowest(0.369 and 0.293).The results showed that Sichuan wheat germplasm was rich in gene diversity and PIC,which was conducive to the exploration of excellent gene resources in breeding.The difference of genetic diversity between landraces and cultivated varieties in Sichuan province was analyzed.It was found that the PIC and gene diversity of all chromosomes of landraces were lower than that of cultivated varieties except chromosome 2A.But the main allele frequency of landraces was higher than that of cultivated varieties.3.Genome-wide association study was carried out to detect the associated markers of target traits.With a threshold of-log10(P)≥3,208 markers were detected in at least two environments,168 markers were detected in three environments.Of these markers,5,36,6,20,5,41,1,2,19,9,6,and 4 significant molecular markers were associated with number of effective tillers,plant height,spike length,awn length,total spikelet per spike,infertile spikelet number,fertile spikelet number,spikelet density,grain number per spikelet,grain number per spike,grain weight per spike,and thousand kernel weight,respectively.Besides,2,8,18,13,and 13 significant molecular markers were associated with grain linear length,grain linear width,grain projection area,grain projection perimeter and grain projection width-length ratio,respectively.According to the LD decay distance of 2.25 Mb,the significantly associated molecular markers detected repeatedly in multiple environments were divided into 173 QTLs,which explained 3.3%-45.4% of the phenotypic variation.4.Haplotype analysis was carried out for potential novel QTL of grain number per spike,total spikelet per spike and thousand kernel weight.It showed that TKW-Hap2 was an excellent haplotype variation to increase the thousand kernel weight and yield of wheat,KNS-Hap2 was an excellent haplotype variation to increase the grain number per spike,and TSS-Hap1 was an excellent haplotype variation to increase the total spikelet per spike.Among the excellent haplotype variation combinations,germplasms containing a combination of TKW-Hap2/TSS-Hap1/KNS-Hap2 showed the higher average yield than that only containing one haplotype variation of TSS-Hap1.Suitable haplotype combinations could be selected for genetic improvement of yield.5.Eleven potential QTLs associated with number of effective tillers,total spikelet per spike,grain number per spike and thousand kernel weight were used for the analysis of candidate genes.Finally,a total of 50 candidate genes were found,which were involved in cell differentiation and proliferation,regulation of plant hormones,carbohydrate transport and synthesis,and amino acid metabolism.