Genome-wide Linkage Mapping And Association Analysis For Processing Quality In Wheat

Wheat processing quality is a quantitative trait controlled by multiple genes. QTL mapping and association analysis are main approaches for studying quantitative traits. However, previous studies were primarily based on simple sequence repeat （SSR） markers, resulting in relatively large genetic distances defining QTL due to the limited number of markers. Therefore, it is difficult for cloning genes and marker-assisted selection in wheat breeding based on the results of QTL mapping. The aim of the present study is to identify QTL and candidate genes for dough strength and starch pasting properties evaluated by Mixograph, Rapid Visco-Analyzer （RVA） and Mixolab parameters using a recombinant inbred line （RIL） population derived from a cross between Gaocheng 8901 and Zhoumai 16 and 247 common wheat cultivars （lines） by genome-wide linkage mapping and association study, respectively and the closely linked SNP markers to QTL can be used in marker-assisted breeding in combination with the KASP technique for improvement of processing quality in bread wheat. The main results were as follows:1. A high-density genetic linkage map was constructed by genotyping 176 RILs derived from the Gaocheng 8901/Zhoumai 16 cross with the wheat 90K iSelect array and 21 functional markers. The genetic linkage map comprised 9,297 SNPs and 7 functional markers, and spanned a total length of 2,789.2 cM involving all 21 chromosomes, with an average chromosome length of 132.8 cM and marker density of 0.30 cM/marker. The A genome included 3,554 SNPs covering a length of 1,013.5 cM, with an average marker density of 0.29 cM/marker; the B genome had 4,961 SNPs covering 1,029.5 cM, with an average marker density of 0.21 cM/marker; the D genome included 789 SNPs, with a length of 746.2 cM, and an average marker density of 0.95 cM/marker. The number of SNP markers in each chromosome ranged from 12 （5D） to 1,222 （5B）; the length ranged from 17.2 cM （6D） to 193.8 cM （4A） and the marker-density ranged from 8.78 cM/marker （5D） to 0.11 cM/marker （2B）. Therefore, SNPs are the most efficient markers for constructing high-density genetic linkage maps.2. Composite interval mapping identified 119 QTL on 20 chromosomes except 4D. Most of QTL were distributed on chromosomes 2B,4B,5A,5B and 6B, explaining from 1.7 to 52.6% of the phenotypic variation. Among them,15 accounted for more than 10% of the phenotypic variation across two or three environments. Compared with previous studies,12 QTL for Mixograph parameters,17 for RVA parameters and 55 for Mixolab parameters were new loci. Eight candidate genes related to dough strength and starch pasting properties were identified. Of them, four genes were involved in biosynthesis of amino acids, two related to starch and sucrose metabolism, and two associated with fatty acid biosynthesis.3. Fifty-nine RILs randomly selected from the Gaocheng 8901/Zhoumai 16 population were genotyped with the wheat 660K iSelect array in order to increase the marker density of the important QTL regions. Additional six markers from the 660K chip were newly mapped to QC3.caas.3AS.2 region and narrowed down the flanking marker interval from 10.0 cM to a 2.0 cM. The distance of the closest marker to the LOD contour peak of QMPV.caas.5AL was shortened from 1.1 to 0.7 cM by the additional marker from the 660K assay. The marker interval flanking QFV.caas.5AS was shortened from 4.6 to 0.3 cM. Additional two markers from the 660K assay were mapped to the QTL cluster on chromosome 4B in the region of 47.4-49.7 cM, and the marker intervals flanking QMTxW.caas.4B, QDT.caas.4B and QST.caas.4BS were reduced from 2.3 to 1.6 cM.4. A GWAS for Mixograph and RVA parameters was performed on 247 common wheat cultivars grown in two environments using 22,468 high quality SNPs, with a minor allele frequency more than 5%and known chromosomal locations, and 15 functional markers based on Q+K mixed liner model. A total of 150 SNPs on 21 chromosomes were identified to be associated with Mixograph and RVA parameters at a significant threshold of -log10 （P）>3, accounting for 4.0～32.7% of the phenotypic variation. Thirty-two stable marker-trait associations （MTAs） were identified, among them,19 SNPs were associated with Mixograph parameters and others were related to RVA parameters. Twenty-three loci for Mixograph parameters and 32 for RVA parameters were reported for the first time compared with previous reports.5. A GWAS for Mixolab parameters was conducted on 156 common wheat cultivars grown in four environments using 18,207 high quality SNPs, with a minor allele frequency more than 5% and known chromosomal locations, and 15 functional markers based on Q+K mixed liner model. A total of 145 SNPs on 20 chromosomes except 4D were found to be associated with Mixolab parameters at a significant threshold of -log10 （P）>3, explaining 7.0～26.3% of the phenotypic variation. Twenty-nine stable MTAs were identified. The GWAS for Mixolab parameters was reported for the first time.6. Fourteen QTL were discovered in both the genome-wide linkage mapping and GWAS simultaneously. The loci on chromosomes 1AL、1BS、1DL、2BS （Ku_c21663₁390）、5AL、5DS、6AS were associated with dough strength and the loci on chromosomes 1BL、2AL、2BS （IACX712）、3AS、4BL、6BL, 7BL were related to starch pasting properties. These loci were valuable for fine mapping, candidate gene cloning and marker-assisted breeding. It was proven that linkage mapping and GWAS can be complementary and validated each other.