SummaryThe marriage of metabolomic approaches with genetic design has proven a powerful tool in dissecting diversity in the metabolome and has additionally enhanced our understanding of complex traits. That said, such studies have rarely been carried out in wheat. In this study, we detected 805 metabolites from wheat kernels and profiled their relative contents among 182 wheat accessions, conducting a metabolite‐based genome‐wide association study (mGWAS) utilizing 14 646 previously described polymorphic SNP markers. A total of 1098 mGWAS associations were detected with large effects, within which 26 candidate genes were tentatively designated for 42 loci. Enzymatic assay of two candidates indicated they could catalyse glucosylation and subsequent malonylation of various flavonoids and thereby the major flavonoid decoration pathway of wheat kernel was dissected. Moreover, numerous high‐confidence genes associated with metabolite contents have been provided, as well as more subdivided metabolite networks which are yet to be explored within our data. These combined efforts presented the first step towards realizing metabolomics‐associated breeding of wheat.

Summary Plants produce numerous metabolites that are important for their development and growth. However, the genetic architecture of the wheat metabolome has not been well studied. Here, utilizing a high‐density genetic map, we conducted a comprehensive metabolome study via widely targeted LC‐MS/MS to analyze the wheat kernel metabolism. We further combined agronomic traits and dissected the genetic relationship between metabolites and agronomic traits. In total, 1260 metabolic features were detected. Using linkage analysis, 1005 metabolic quantitative trait loci (mQTLs) were found distributed unevenly across the genome. Twenty‐four candidate genes were found to modulate the levels of different metabolites, of which two were functionally annotated by in vitro analysis to be involved in the synthesis and modification of flavonoids. Combining the correlation analysis of metabolite‐agronomic traits with the co‐localization of methylation quantitative trait locus (mQTL) and phenotypic QTL (pQTL), genetic relationships between the metabolites and agronomic traits were uncovered. For example, a candidate was identified using correlation and co‐localization analysis that may manage auxin accumulation, thereby affecting number of grains per spike (NGPS). Furthermore, metabolomics data were used to predict the performance of wheat agronomic traits, with metabolites being found that provide strong predictive power for NGPS and plant height. This study used metabolomics and association analysis to better understand the genetic basis of the wheat metabolism which will ultimately assist in wheat breeding.

Genome-wide association studies (GWAS) have been widely used to dissect the complex biosynthetic processes of plant metabolome. Most studies have used single-locus GWAS approaches, such as mixed linear model (MLM), and little is known about more efficient algorithms to implement multi-locus GWAS. Here, we report a comprehensive GWAS of 20 free amino acid (FAA) levels in kernels of bread wheat (Triticum aestivum L.) based on 14,646 SNPs by six multi-locus models (FASTmrEMMA, FASTmrMLM, ISISEM-BLASSO, mrMLM, pKWmEB, and pLARmEB). Our results showed that 328 significant quantitative trait nucleotides (QTNs) were identified in total (38, 8, 92, 45, 117, and 28, respectively, for the above six models). Among them, 66 were repeatedly detected by more than two models, and 155 QTNs appeared only in one model, indicating the reliability and complementarity of these models. We also found that the number of significant QTNs for different FAAs varied from 8 to 41, which revealed the complexity of the genetic regulation of metabolism, and further demonstrated the necessity of the multi-locus GWAS. Around these significant QTNs, 15 candidate genes were found to be involved in FAA biosynthesis, and one candidate gene (TraesCS1D01G052500, annotated as tryptophan decarboxylase) was functionally identified to influence the content of tryptamine in vitro. Our study demonstrated the power and efficiency of multi-locus GWAS models in crop metabolome research and provided new insights into understanding FAA biosynthesis in wheat.