Skip to main content
Download PDF

Metabolic-GWAS provides insights into genetic architecture of seed metabolome in buckwheat

BMC Plant Biology volume 23, Article number: 373 (2023) Cite this article

Abstract

Background

Buckwheat (Fagopyrum spp.), belonging to the Polygonaceae family, is an ancient pseudo-cereal with high nutritional and nutraceutical properties. Buckwheat proteins are gluten-free and show balanced amino acid and micronutrient profiles, with higher content of health-promoting bioactive flavonoids that make it a golden crop of the future. Plant metabolome is increasingly gaining importance as a crucial component to understand the connection between plant physiology and environment and as a potential link between the genome and phenome. However, the genetic architecture governing the metabolome and thus, the phenome is not well understood. Here, we aim to obtain a deeper insight into the genetic architecture of seed metabolome in buckwheat by integrating high throughput metabolomics and genotyping-by-sequencing applying an array of bioinformatics tools for data analysis.

Results

High throughput metabolomic analysis identified 24 metabolites in seed endosperm of 130 diverse buckwheat genotypes. The genotyping-by-sequencing (GBS) of these genotypes revealed 3,728,028 SNPs. The Genome Association and Prediction Integrated Tool (GAPIT) assisted in the identification of 27 SNPs/QTLs linked to 18 metabolites. Candidate genes were identified near 100 Kb of QTLs, providing insights into several metabolic and biosynthetic pathways.

Conclusions

We established the metabolome inventory of 130 germplasm lines of buckwheat, identified QTLs through marker trait association and positions of potential candidate genes. This will pave the way for future dissection of complex economic traits in buckwheat.

Background

Buckwheat (Fagopyrum spp.) is a pseudo-cereal belonging to the Polygonaceae family. The genus Fagopyrum contains 27 species, of which two diploid species—Tartary buckwheat (Fagopyrum tataricum) and common buckwheat (Fagopyrum esculentum) are grown for food [1]. These two species vary in their breeding system, F. tataricum being self-fertile mainly perform inbreeding, while F. esculentum is an insect-pollinated, and obligate out breeder [2]. The grain is consumed after boiling or steaming, or powdered into gluten-free flour.

Buckwheat is a valuable health-promoting crop that is used for preparing number of functional foods and nutraceuticals products. It has high-quality, gluten-free proteins as well as a wealth of bioactive ingredients and antioxidants. These characteristics account for its current high demand [3]. The balanced essential amino acids, resistant starch, vitamins, and minerals available in buckwheat are beneficial to human health. Additionally, it is a rich source of bioactive flavonoids like epicatechin, rutin, quercetin, and (iso)vitexin, all of which have been shown to have favourable effects on hyperlipidaemia, diabetes, and vascular diseases [4,5,6]. In particular, Tartary buckwheat is abundant in rutin, a citrus flavonoid that helps the body to use vitamin C, produce collagen and strengthen blood vessels [7].

In view of the increasing interest in nutraceutical crops as potential candidates for crop diversification and a shift from food to nutritional and health security, a large scale nutri-genomic investigations across crops has been initiated, in order to unravel the potential of underutilised species as frontline foods to promote health [8]. Despite its obvious potential as a functional food, buckwheat has not been fully harnessed due to low yield, self-incompatibility, increased seed cracking, limited seed set, lodging, and frost susceptibility. To overcome these bottlenecks, there is a need to improve these traits in this underutilized species [9]. The availability of diverse data, as well as the population structure of crop germplasm, will be valuable genetic resources for discovering genes in buckwheat for improving it as a potential crop for the future [10].

In-depth research on the flavonoid biosynthesis pathway in model plants and several crop species has recently gained attention. These available experimental evidences indicate that metabolomics can provide novel insights into the biosynthesis pathways, especially in crops that possesses high value traits associated with bioactive antioxidant metabolites. Buckwheat being well known for such traits is attracting high throughput omics research in the recent years. Metabolome Based Genome Wide Association Study (mGWAS) or metabolomic quantitative trait loci (mQTL) mapping, in particular, is emerging as a powerful tool for mining the genetic loci contributing to metabolite diversification. Furthermore, the relationship of these metabolites, naturally deputed to defence against biotic and abiotic stress [11] with food quality and flavour has been thoroughly investigated [12]. However, to the best of our knowledge, no mGWAS or mQTL studies in buckwheat have been reported till date. With the advancement of genome sequencing and bioinformatics technologies, approaches such as association and linkage mapping have come a long way to unravel the genetic diversity of targeted traits across crops. Furthermore, plant association mapping has revealed new genetic and biochemical information about metabolomes. Genome-wide association studies (GWAS) are frequently used to identify new genes and QTLs by locating significant allelic differences in candidate genes underpinning quantitative and complicated traits, such as those linked to growth, development, stress tolerance, and nutritional quality [13]. As a potential technique for moving forward, mGWAS entails merging genotyping and metabolome data from the diverse crop germplasm [14], with a striking example from highland barley (Qingke), mGWAS study has been carried out for mining genes involved in phenylpropane metabolic pathway [15].

Considering no such information is available till date on buckwheat, in the present study, high throughput metabolomics analysis was carried out on the seed endosperm of 130 diverse buckwheat genotypes and GBS-based SNP genotyping was performed to identify marker trait associations. Further, candidate genes located in 100 kb region of known QTLs were identified.

Results

Identification of phenolic metabolites in buckwheat extracts

Results obtained from HPLC–DAD-MS characterisation are reported in Supplementary Table 1, and a representative chromatogram is also shown (Supplementary Fig. 1). Among the identified metabolites, 10 were flavonoids (both glycosidic derivatives and aglycones), 6 were phenolic acids (benzoic, caffeic, and ferulic acids derivatives), 3 were catechin derivatives, and 3 were gallic acid derivatives. Among the flavonoids, Rutin and its aglycone (quercetin) possess numerous biological activities. Exposure of buckwheat grain to moisture results in enzymatic breakdown of rutin to quercetin by rutinosidase, as it also happens after milling and mixing of the flour with water. The beneficial effect of rutin depends upon its concentration in the final product that further depends upon the conversion rate of rutin to quercetin. Thus, the quantification of both rutin and quercetin are considered at seed level. Out of 22 different metabolites, 18 were found to be significantly associated with different SNP markers (Table 1).

Table 1 Details of significant QTLs associated with seed metabolome content
Full size table

Metabolite fingerprinting of buckwheat samples

Raw HPLC–DAD-MS data were analysed using multivariate statistics. An explorative analysis was performed, namely PCA. Results, reported in Fig. 1a, showed a clear distinction between F. tataricum and F. esculentum samples, indicating different metabolome compositions. The same group’s distinction can be observed from the heatmap in Fig. 1c. PCA score plot revealed that the variance explained by PC1 is higher than that explained by PC2 (26.6% vs. 14.8%), indicating that the inter-group variability is higher than the intra-group one.

Fig. 1
figure 1

Metabolomic analysis of buckwheat samples. a PCA score plot. Red dots: F. esculentum; green dots: F. tataricum; (b) Volcano plot showing variables significantly associated to the two buckwheat sample groups. Variables on the right side of the plot are more abundant in F. tataricum, while those on the left side are more abundant in F. esculentum. Only variables significantly (FDR-p < 0.05) associated to one of the two groups are highlighted; (c) Heatmap plot. Red: F. esculentum; green: F. tataricum; (d) Boxplots showing the comparison of the amounts of significant variables in the two groups of buckwheat samples. Red: F. esculentum; Green: F. tataricum

Full size image

Variables significantly associated with the two sample groups were selected using a volcano plot, where variables with FDR-adjusted p-value < 0.05 and with Fold Change > 2 were considered as significant descriptors. As can be observed in Fig. 1b, four variables were significantly associated to the F. tataricum samples group, namely rutin, quercetin, kaempferol and epiafzelchin-epicatechin-O-dimethylgallate (EED). The amounts of these variables in the two groups of samples are shown in Fig. 1d.

Characterization and distribution of SNPs in buckwheat

The data from the sequencing platform (with a sequencing depth of 121x) represent an average of 1.58 million reads per sample with a read length of 150 base pairs. The reads were then mapped to the buckwheat reference genome (GCA 002319775.1; http://www.mbkbase.org/Pinku1/) with an average mapping percentage of 90.78. The mapping of reads resulted in the identification of 4,142,684 variants, containing 3,728,028 SNPs and 414,656 InDels (214,798 insertions and 199,858 deletions). However, while considering 5% minor allele frequency and 30% missing rate, a total of 34,978 SNPs were observed. The chromosome wise distribution of SNP is shown in Fig. 2. The highest number of filtered SNPs (6750) was observed on chromosome (chr.) 1, whereas the lowest number of SNPs (3190) was found on chr. 7. SNPs have also been classified as having a high, low, moderate, or modifier impact, with the percentage of high being 0.413%, low being 1.566%, moderate being 2.105%, and modifier being 95.916%. According to the effects by functional class missense percent was 59.93%, nonsense was 3.37% and silent was 36.70%. Number of effects by type and region are mentioned in detail in Supplementary Table 2. Total number of transitions and transversions were found to be 39,416,882 and 23,590,338 respectively, with a transition by transversion ratio (Ts/Tv) of 1.6709.

Fig. 2
figure 2

Chromosome wise SNP distribution in buckwheat germplasm

Full size image

In order to ensure similarity between two genomes of buckwheat, we performed pairwise alignment of both genomes using the GSALIGN program and visualized the result using DotPlot. Surprisingly from genome alignment, the major scaffolds from the common buckwheat genome aligned strongly with the Tataricum genome (Supplementary Fig. 2). So, we concluded that utilizing the Tataricum genome as a reference genome for annotating traits of common buckwheat doesn’t drift the overall study. It was further validated by using the common buckwheat genome (https://doi.org/10.1111/jipb.13459) [16] as reference genome to revalidate the already identified SNPs. We found a total of 68 QTLs (~ 67%) common among both genomes as detailed in Table 1. Moreover 31QTLs have the same residues in Fagopyrum esculentum genome as the genotypes under study.

Genetic diversity and population structure

All paired genetic distances between the 130 buckwheat lines in this study were calculated using SNP-based genotypic data. A neighbouring tree revealed that the genotypes were divided into four main groups, which were further subdivided based on the genetic distances (Fig. 3a). Dendrogram analysis revealed that among the four major groups there was one minor group which was clustered together and included only three genotypes, i.e. BWM30, BWM38 and BWZ49. These three genotypes were all from the same species, F. esculentum, and are thus closely related. PCA revealed variations among buckwheat genotypes (Fig. 3b). Furthermore, the population structure was scored for K values ranging from 1 to 12 across the panel using high quality SNPs in population structure analysis. The delta K peak was found to be the highest at K = 4, and 130 buckwheat genotypes were classified into four populations (Fig. 3c and d). Furthermore, this was consistent with the neighbour-joining tree with only minor deviations.

Fig. 3
figure 3

SNP markers based population analysis. a UPGMA dendrogram showing genetic relationship among 130 genotypes; (b) PCA Plot of Buckwheat genotypes; (c) Peak of delta K; (d) structure analysis indicated genotypes grouping into four sub-populations based on membership coefficients indicated on vertical coordinate

Full size image

Marker trait association

A total of 99 markers were found to be having significant association with 18 metabolites. The details of these marker trait associations are summarized in Table 1 and depicted in Manhattan and QQ-Plots (Fig. 4 A-R). GWAS was performed for 18 buckwheat seed metabolites (catechin, orientin, rutin, quercetin, EAEC, vanillin, GEDC, duratin, keampferol, QDG, DH, EED, SI, EEMG, FARD, CAH, UNK2 and CG). The analysis was done using GAPIT CMLM method. GAPIT uses FDR adjusted p-value to filter significant SNPs. This approach reduces chances of false positive SNP makers in further analysis. In Manhattan plot dotted green line shows p-value and solid green line shows FDR adjusted p-value. Out of total 3,728,028 SNPs, 34,978 were found to be significantly associated with different seed metabolites with 12.14–32.04% phenotypic variance. A total of 27 SNPs were found significantly associated with the metabolite ferulic acid rhamnosyl derivative (FARD). Out of 27, 12 SNPs were found in chr. 1; as such this region could be regarded as hot-spot of SNPs for this particular metabolite. One SNP on chr.6 (p-value = 8.44E-07) contributed for 32.04% of phenotypic variation. For metabolite epiafzelchin epicatechin-o-dimethylgallate (EED), 11 SNPs were significantly associated. These were positioned on chr.1, chr.2, chr.4, chr.5, and chr.7 with the highest number of SNPs (5) on chr.1 and the lowest (1) on chr.2, chr.5 and chr.7, respectively. One SNP on chr.1 positioned at 63,055,659 (p-value = 4.62E-05) contributed 15.05% to phenotypic variation. Ten SNPs were found associated with quercetin on each of chr.2, chr.4, chr.5, chr.7, and chr.11. One SNP on chr.5 positioned at 29,043,577 (p-value = 1.01E-05) contributed 21.64% phenotypic variation. For vanillin, 7 SNPs were found significantly associated that are positioned on chr.1, chr.2, chr.4, chr.5 and chr.7. One SNP on chr.4 positioned at 37,569,986 (p-value = 3.56E-05) contributed 23.24% phenotypic variation. For catechin, 7 SNPs were significantly associated and are positioned on chr.1, chr.3, chr.4, chr.5 and chr.6. One SNP on chr.3 positioned at 5,955,359 (p-value = 8.79E-06) contributed 23.28% phenotypic variation. For diacaffeoyl-hexoside (DH), 5 SNPs were found significantly associated and are positioned on chr.6 and chr.7. One SNP on chr.6 positioned at 31,943,608 (p-value = 1.85E-05) contributed 15.87% phenotypic variation. For swertiamacroside isomer (SI), 5 SNPs were found significantly associated that are positioned on chr3, chr4 and chr5. One SNP on chr.5 positioned at 9,638,542 (p-value = 6.62E-05) contributed 15.30% phenotypic variation. For quercetin 3-D-glucoside (QDG), 4 SNPs were found significantly associated that are positioned on chr.1, chr.6 and chr.7. One SNP on chr.7 positioned at 36,838,233 (p-value = 8.32E-06) contributed 17.41% phenotypic variation. For duratin, 4 SNPs were found significantly associated that are positioned on chr.1, chr.3 and chr.5. One SNP on chr.1 positioned at 41,194,007 (p-value = 3.84E-05) contributed 25.26% phenotypic variation. For galloyl ester of 5,6,7-trihydroxy-2,3 dihydrocyclopents(b)chromene-1,9-dione-3-carboxylic acid hexoside (GEDC), 3 SNPs were found significantly associated that are positioned on chr.1, chr.3 and chr.7. One SNP on chr.1 positioned at 19,115,204 (p-value = 1.02E-04) contributed 15.81% phenotypic variation. For orientin, 3 SNPs were found significantly associated that are positioned on chr.5 and chr.8. One SNP on chr.8 positioned at 44,828,457 (p-value = 3.37E-05) contributed 20.15% phenotypic variation. For catechin glycoside (CG), 3 SNPs were found significantly associated that are positioned on chr.3 and chr.6. One SNP on chr.3 positioned at 5,955,778 (p-value = 1.56E-04) contributed 14.75% phenotypic variation. For caffeic acid hexoside (CAH), 3 SNPs were found significantly associated that are positioned on chr.1, chr.2 and chr.3. One SNP on chr.3 positioned at 27,868,660 (p-value = 9.36E-05) contributed 14.70% phenotypic variation. For Rutin, 2 SNPs were found significantly associated that are located on chr3 and chr5. and contributed 12.70% and 15.66% phenotypic variation. For kaempferol, 2 SNPs were found significantly associated that are located on chr.1 and chr.5 and contributed 17.69% and 19.87% phenotypic variation. For (epi)afzelchin-(epi) catechin (EAEC), only 1 SNP was found significantly associated that is located on chr.5 and contributed 26.40% to phenotypic variation. For epiafzelchin epicatechin-o-methylgallate (EEM), only 1 SNP was found significantly associated that is located on chr.1 and contributed 23% to phenotypic variation. For UNK2, only 1 significantly associated SNP was found on chr.2 and contributed 17.11% to phenotypic variation.

Fig. 4
figure 4

The figure showing the Manhattan plots and Q-Q plots of different metabolites. a Catechin, (b) Orientin, (c) Rutin, (d) Quercetin, (e) (epi)afzelchin-(epi)catechin, (f) Vanillin, (g) Galloyl ester of 5,6,7-trihydroxy-2,3 dihydrocyclopents (b)chromene-1,9-dione-3-carboxylic acid hexoside, (h) Duratin, (i) Keampferol, (j) Quercetin, 3-D-glucoside, (k) Diacaffeoyl-hexoside, (l) Epiafzelchin epicatechin-o-dimethylgallate, (m) Swertiamacroside isomer, (n) Epiafzelchin epicatechin-o-methylgallate, (o) Ferulic acid rhamnosyl derivative, (p) Caffeixc acid hexose, (q) UNK2 and (r) Catechin Glycoside

Full size image

LD Plot and haplotype blocks

LD was calculated from 4,142,684 pairs using 100 markers sliding window operation, out of which 8% was with zero LD and 23% was found in the significant range (p-value < 0.05). As the physical distance increases, the r2 distribution showed a rapid LD decay for all genotypes. A total of 1783 haplotype blocks were identified, containing 68% markers (Supplementary Figs. 3 and 4). The blocks were uniformly distributed among all chromosomes.

In-silico analysis for candidate gene identification

In-silico analysis revealed a total of 168 genic sequences linked with different marker traits (Table 1). The candidate genes were subjected to pathway analysis using the KEGG-KASS server and GO process identification using the Uniprot database to gain insight into the biological process. According to GO analysis, the key biological processes involved are biotic stimulus response, phospholipid biosynthetic process, protein phosphorylation, lipid transport, oxidative stress response, and ion transport (Table 2). The key molecular functions of the identified candidate genes were flavin adenine dinucleotide binding, cysteine-type peptidase activity, protein hetero dimerization and protein binding (Table 2). KEGG-KASS server revealed that the identified candidate genes were related to metabolic pathways of butanoate, glycerol-phospholipids, arachidonic acid, glutathione, alanine, aspartate and glutamate and biosynthetic pathways of secondary metabolites such as flavonoids, sesquiterpenoids and triterpenoids (Table 3).

Table 2 Gene ontology annotation of the identified candidate genes
Full size table
Table 3 The description of identified genes related to different metabolic pathways
Full size table

Discussion

The enormous diversity of structurally distinct metabolites found in the plant metabolome are genetically controlled. It has been hypothesised that species-level metabolome variations are significantly more extensive than previously believed [17]. This calls for the integration of metabolomics and genetics techniques like QTL and GWAS for examining the genetic control of the metabolome and enabling the delineation of metabolic pathways and the dissection of agronomic features [18,19,20]. Most of the reported metabolites are necessary for the plant's survival, as well as for its ability to grow and interact with its environment [21]. Some of these metabolites confer special nutritional benefits to crops like buckwheat [22]. Our endeavour to characterise buckwheat secondary metabolites was based on the premise that these compounds provide an effective way of defence against biotic and abiotic challenges as well as contribute to the nutritional quality of this valuable crop [23, 24].

Metabolomic profiling has been frequently used in conjunction with genetic techniques like genome-wide association studies (GWAS as mGWAS) and quantitative trait loci (QTL as mQTL) to discover the functional genes behind the variation in metabolite content of different plant species. Tomato and Arabidopsis were the first plants wherein mQTL tools were applied previously [25, 26]. Following these ground-breaking researches, the mQTL technique was widely modified to identify the various genetic components controlling the metabolome in many plant species including the Tartary buckwheat [27, 28], offering insights into the genetic and biochemical underpinnings of metabolic pathways. An obvious advantage of using metabolomics is the complexity of phenotyping as well as difficulties associated with large scale field phenotyping. The relative levels of numerous metabolites can be comprehensively profiled with ease and associated to the phenotypes of interest either directly or indirectly [29]. Therefore, identifying putative functional genes governing the variation in metabolite concentration may aid in our understanding of important crop characteristics. A huge population's genes can be identified using GWAS at substantially higher mapping resolutions [14]. In rice, the metabolite trigonelline (N-methyl nicotinic acid) implicated in grain width with its underlying genetic factors were discovered using the same approach.

The current study focuses on the use of mGWAS to identify candidate genes and metabolic pathways in buckwheat. Metabolic pathways are composed of highly varied yet vaguely linked metabolites, which could be thought of as chemical decorations on a number of fundamental structures [30]. Here we not only reported metabolomic profiling data from buckwheat seed but also discovered high-confidence candidate genes. The ultimate objective of this study is the understanding of the genetics underlying traits of interest, which in turn may benefit breeding efforts that seek improvement in economically important traits in buckwheat. The current study used genome-wide association to identify 99 significant markers underlying the studied traits. We were able to predict important genes that encode metabolic and biosynthetic pathways. An enriched gene involved in the flavanoid biosynthesis pathway was also discovered. These findings are significant because flavonoids are the largest class of secondary metabolites found in plants and have a wide range of functional roles, including pigments and antioxidant properties [30]. In addition to flavonoids, different components of amino acid metabolism pathways of alanine, aspartate, glutamate, glutathione, valine, leucine and isoleucine were identified. These defence compounds help plants to withstand various stresses such as hypoxia, water logging, and drought, and act indirectly as a precursors to the compounds pantothenate and CoA, as well as play a central role in amino acid metabolism. They have been implicated in physiological response to stress, normal metabolism and energy processes such as photosynthesis and respiration. Similarly, glutathione is crucial for the regulation of cellular processes, nutrition metabolism, and antioxidant defence. To that end, our data not only provides specific candidate genes as molecular resources that can be effectively used after validation, but also enlightens further metabolite network exploration (Supplementary Figs. 5 and 6). The information generated in the present study will surely facilitate the metabolomics-associated breeding of buckwheat in the future.

Conclusion

The present study exemplified the potential of integrating GBS technology with metabolomics that led to the discovery of a significant number of potential SNP markers for association mapping and is a valuable resource for QTL studies for the breeding programmes. As such, 68 common QTLs identified in this study by using F. tataricum and F. esculentum reference genomes might have better implications in improving metabolome content in both common and Tartary buckwheat through molecular breeding approach.

Additionally, the identified candidate genes with potential roles can be explored further through more extensive research. The findings of this study will promote the efficient use of genetic and genomic resources aimed to raising the yield potential and enhancing the metabolite contents and overall quality of buckwheat.

Methods

Plant material

A total of 130 diverse buckwheat genotypes were used as plant material in the present study (Supplementary Table 3). Buckwheat germplasm was collected from different geographical regions of Western Himalayan state of Jammu and Kashmir, India and some of the genotypes were also procured from National Bureau of Plant Genetic Resources (NBPGR), New Delhi (Supplementary Fig. 7). Most of the collected genotypes were maintained in the crop research fields of Sher-e-Kashmir University of Agricultural Sciences & Technology of Kashmir, India.

Quali-quantitative characterization of phenol secondary metabolites

Phytochemical characterization of buckwheat samples was performed by HPLC–DAD-MS. Samples were prepared by extracting 200 mg of dried powdered samples in 25 mL of a 50:50 v/v methanol:water mixture, using an ultrasound bath to increase the extraction efficiency (20 min at r.t.). After centrifugation at 13,000 rpm for 10 min, supernatant was collected and directly injected in the instrument for analysis.

Chromatographic separation was performed using a Phenomenex Synergy MAX-RP 80A (4 µm, 150 × 2.0 mm) column as stationary phase, and a mixture of 1% formic acid in water (A) and acetonitrile (B) as mobile phase. Elution gradient was set to the following setting: 0 min, 95% A; 10 min, 50% A; 13 min, 50% A; 18 min, 10% A; 19 min, 10% A; 20 min, 95% A. Column was left to equilibrate for 5 min. Flow rate was 0.4 mL/min with injection volume set to10 µL.

Identification of eluted compounds was performed using integrated DAD and MS data. UV–Vis absorbance was monitored in the range 200–600 nm, and the spectrum of each eluted compound was used to determine its chemical class. Regarding MS, fragmentation data of each compound obtained from MSn experiments were compared with literature data to identify the eluted compounds. The following conditions was set for MS: needle voltage- 4500 V; capillary voltage- 70 V; RF loading- 100%; nebulising gas pressure- 20 psi (nitrogen); drying gas pressure- 15 psi; drying gas temperature- 350 °C. and mass range was 50–2000 Da. Fragmentation patterns of eluted compounds were obtained using the turbo detection data scanning (TDDS®) function of the instrument, setting n = 4 levels of fragmentation.

For the quantification of phenolic compounds from buckwheat samples, standard calibration curves built from DAD measurements were used. Flavonoids were quantified using rutin as standard compound (1-100 µg/mL in methanol), and calibration curve (y = 51.52x – 183.22; R2 = 0.999) was built by monitoring the absorbance values of standard solutions at λmax = 350 nm. For phenolic acids, chlorogenic acid was used as reference compound (0.9–90 µg/mL in methanol), and the calibration curve (y = 90.54x – 32.84; R2 = 0.999) was built monitoring the absorbance values of standard solutions at λmax = 280 nm. Standard solutions (1–100 µg/mL in methanol) of gallic acid were used for the quantification of gallic derivatives, and the calibration curve was built at λmax = 280 nm: y = 109.72x – 68.33; R2 = 0.999. Finally, catechin derivatives were quantified using a catechin calibration curve (y = 20.81x – 29.61; R2 = 0.999) and was built analysing the absorbance value at λmax = 280 nm of catechin solutions (methanol) in the concentration range of 1–100 µg/mL.

Metabolomics analysis

Metabolomics exploration of buckwheat samples dataset was performed using the Metaboanalyst v. 5.0 platform [31]. For this, quali-quantitative chemical data were organized in a proper data matrix and submitted to the web platform. Data were log transformed and Pareto scaled before analysis. This was accomplished through the use of both unsupervised Principal Component (PCA) and heatmap analyses, as well as supervised methods such as Partial Least Squares Discriminant Analysis (PLS-DA). To avoid over fitting of results, PLS-DA models were validated by using both permutation test (1000 random permutations) and leave-one-out cross validation, whose R2, Q2 and accuracy parameters were used to assess the robustness and predictability of the models. Variables (metabolites) significantly associated to specific sample groups were selected by using a Volcano plot, setting as threshold values FDR-adjusted p-value < 0.05 and Fold Change > 2.

DNA extraction, library preparation and sequencing

Seeds of 130 diverse genotypes of buckwheat was sown in plastic trays for three weeks in a polyhouse and the harvested shoots were used for genomic DNA extraction using CTAB method and the quality as well as quantity of DNA was checked on both gel electrophoresis (0.8% Agarose) and nano-drop (mySPEC, Wilmington, USA). GBS libraries were prepared following the method reported in [32], with minor modification. 20 µL digestion reaction contained 1X NEB Buffer, 3.6 U ApeKI and 100 ng of DNA was digested for 4 h at 75 °C. The barcoded adapters were then ligated to sticky ends by using T4 ligase (New England Biolabs). To inactivate the T4 ligase, samples were incubated at 22 °C for 1 h before being heated to 65 °C for 30 min. The sets of 130 digested DNA samples were combined (5 µL each), each with a different barcode adapter, and purified using a commercial kit (QIAquick PCR Purification Kit; Qiagen, Valencia, CA) according to the manufacturer's instructions. DNA samples were eluted in a 25 µL final volume.

Restriction fragments from each library were then amplified in 50 µL volumes containing 10 µL pooled DNA fragments, 25 µl of KAPA HiFi Hot Start Ready Mix PCR, and 1 µL each of the P5 and P7 dual indexing primers (12.5 pmol). These primers included complementary sequences for priming future DNA sequencing reactions, attaching PCR products to oligonucleotides that coat the Illumina sequencing flow cell, and amplifying restriction fragments with ligated adapters. With 0.9X AMPure XP beads (Catalog: A63881, Beckman Coulter), the final PCR products were purified to get rid of any primers that weren't used. The final 130-plex DNA library that had been purified was measured using an Agilent Bioanalyzer before being sequenced on an Illumina HiSeqTM X10 platform (Illumina® Inc., San Diego, CA, USA) using V4 sequencing chemicals.

Post-sequencing analysis

The raw reads were filtered for adapter sequences, low quality reads and low-quality residues towards 5` region of the sequence. After quality filtering and data de-multiplexing, the high-quality sequences were mapped to the Tartary buckwheat reference genome assembly (GCA 002319775.1; http://www.mbkbase.org/Pinku1/) using BWA program V 0.7.5 [33]. SNPs were mined from the coding and non-coding regions and were subsequently annotated. The SNPs were annotated to the genic, intergeneic, non-coding and regulatory regions using SNPEFF program [34].

Moreover, a comprehensive comparison of the genetic sequences at the genomic level between F. esculentum (Common Buckwheat) and F. tataricum (Tartary Buckwheat) was performed through pairwise genome alignment, using GSALIGN program (https://github.com/hsinnan75/GSAlign). This process aimed to elucidate the shared characteristics and distinctions within the genomes of these two buckwheat species. The pairwise genome alignment between the two buckwheat genomes encompassed a series of steps, ensuring accurate and reliable results. Initially, the genomic data of both species underwent a pre-processing stage to eliminate any extraneous elements that might introduce noise and potentially hinder the alignment process. By reducing unwanted artefacts, the subsequent alignment was enhanced, allowing for more precise comparison of the genetic sequences. To optimize the alignment, GSALIGN tends to maximise the similarity between corresponding regions while minimizing any gaps that might occur in the alignment. By strategically aligning the sequences, the software facilitated the identification and comparison of specific genetic elements shared between the two species. The results were visualized using DotPlot (https://dotplot.soft112.com/).

Population structure analysis

Population structure was estimated using a Bayesian Markov Chain Monte Carlo model (MCMC) implemented in STRUCTURE v2.3.4 [35]. The filtered SNPs were converted to structure format using PGD Spider version 2.1.1.5. For each population (k) set number from 2 to 7, three runs were completed. For each run, the burn-in period and the MCMC replication number were set to 100,000 and 300,000, respectively. Structure Harvester used the log probability of the data [LnP(D)] and delta K (K) based on the rate of change in [LnP(D)] between subsequent populations to estimate the most likely K-value [36]. The neighbour-joining tree was built using Phylip and MEGA5 [37].

Principal component analysis

PCA was calculated using PLINKV 1.9 [38] and then plotted by using R program. Dendrogram analysis was done using TASSEL V4 using Neighbour-Joining method and then plotted with Structure Q-matrix using iTOL. PCA plot was made on four populations which were detected using Structure. Using high quality SNPs, the population structure was graded for K-values ranging from 1 to 12 across the panel.

Marker trait association

GAPIT V3, an R package that conducts a Genome-Wide Association Study (GWAS) and genome prediction, was used to implement the Compressed Mixed Linear Model (CMLM) [39]. Modern statistical genetics tools including the unified mixed model, EMMA, compressed mixed linear model, and P3D/EMMAx are used in this application. SNPs were considered significant using threshold log10 (p-value) < 1E-4. Manhattan plots and quantile–quantile (QQ) plots were developed using R-package QQMAN. Manhattan plots revealed statistically significant associated markers, and quantile–quantile (QQ) plots were created to graphically depict the associated marker distribution pattern. GAPIT was used to calculate the R squared values (r2) for markers; the r2 value explains the proportion of phenotypic variation explained by each SNP locus.

LD plots and haplotype blocks

Linkage disequilibrium (LD) was measured by the parameter r2 using SNPs with high confidence. The values were calculated using TASSEL v5.0 and the values were plotted against genetic distance (in bp) in R software [40]. A threshold of r2 = 0.2 was used to determine LD extent. The size of LD blocks was determined by fitting the second LOESS decay curve to the r2 values plotted against the physical distance among markers. Using the Gabriel et al. 2002 [41] described confidence interval; Haploview 4.2 was used to identify haplotype blocks from the entire set of SNPs [42]. The analysis excluded heterozygous loci.

Candidate gene identification

The gene containing the SNP was used to determine the probable candidate gene search from the significant SNP-trait associations obtained from mGWAS using the SNPEFF programme V5.1 against F. esculentum annotation downloaded from NCBI [34]. The candidate genes were mapped to the Kyoto Encyclopaedia of Genes and Genomes (KEGG) database using the KEGG-KAAS (KEGG Automatic Annotation Server) server for pathway analysis and Gene Ontology (GO) annotation was carried out using standalone BLASTP and BLASTX [43] against the Uniprot database (release 2022_02) to gain insight into the functional role of candidate genes with SNPs (UniProt Consortium, 2019).

Availability of data and materials

The datasets used and analysed during the current study will be available from the corresponding author on reasonable request.

Abbreviations

GBS:

Genotyping-by-Sequencing

GAPIT:

Genome Association and Prediction Integrated Tool

SNPs:

Single Nucleotide Polymorphisms

QTLs:

Quantitative Trait Loci

mGWAS:

Metabolome Based Genome Wide Association Study

mQTL:

Metabolomic Quantitative Trait Loci

HPLC–DAD-MS:

High-Performance Liquid Chromatography coupled with Diode-Array Detection and electro-spray ionization tandem mass spectrometry

PCA:

Principal Component Analysis

FDR:

False Discovery Rate

LD:

Linkage disequilibrium

KEGG:

Kyoto Encyclopedia of Genes and Genomes

GO:

Gene Ontology

PLS-DA:

Partial Least Squares Discriminant Analysis

References

  1. Zhou M, Ivan K, Sun HW, Nikhil KC, Wieslander G. Molecular Breeding and Nutritional Aspects of Buckwheat; Academic Press: Pittsburgh, PA, USA. 2016;203–207.

  2. Hunt HV, Shang X, Jones MK. Buckwheat: a crop from outside the major Chinese domestication centres? A review of the archaeobotanical, palynological and genetic evidence. Veget Hist Archaeobot. 2018;27:493–506.

    Google Scholar 

  3. Bashir E, Mahajan R, Mir RA, Dar WA, Zargar SM. Unravelling the genetic variability and population structure of buckwheat (Fagopyrum spp): a collection of north western Himalayas. The Nucleus. 2021;64:93–101.

    Article  CAS  Google Scholar 

  4. Joshi DC, Zhang K, Wang C, Chandora R, Khurshid M, Li J, et al. Strategic enhancement of genetic gain for nutraceutical development in buckwheat: a genomics-driven perspective. Biotechnol Advance. 2020;39:107479.

    Article  CAS  Google Scholar 

  5. Ren Y, Wu S, Xia Y, Huang J, Ye J, Xuan Z, Li P, Du B. Probiotic-fermented black tartary buckwheat alleviates hyperlipidemia and gut microbiota dysbiosis in rats fed with a high-fat diet. Food Funct. 2021;12(13):6045–57.

    Article  CAS  PubMed  Google Scholar 

  6. Lu CL, Zheng Q, Shen Q, Song C, Zhang ZM. Uncovering the relationship and mechanisms of Tartary buckwheat (Fagopyrum tataricum) and Type II diabetes, hypertension, and hyperlipidemia using a network pharmacology approach. Peer J. 2017;21(5):e4042.

    Article  Google Scholar 

  7. Zhang L, Li X, Ma B, Gao Q, Du H, Han Y, Li Y, Cao Y, Qi M, Zhu Y, Lu H, Ma M, Liu L, Zhou J, Nan C, Qin Y, Wang J, Cui L, Liu H, Liang C, Qiao Z. The Tartary Buckwheat Genome Provides Insights into Rutin Biosynthesis and Abiotic Stress Tolerance. Mol Plant. 2017;10:1224–37.

    Article  CAS  PubMed  Google Scholar 

  8. Mir RA, Nazir M, Sabreena, Naik S, Mukhtar S, Ganai BA, Zargar SM. Utilizing the underutilized plant resources for development of life style foods: Putting nutrigenomics to use. Plant Physiol Biochem. 2022;171:128–38.

    Article  CAS  PubMed  Google Scholar 

  9. Mukhtar S, Bashir Z, Mir RA, Zargar SM. Genomic Approaches for the Improvement and Conservation of Buckwheat. In Neglected and Underutilized Crops-Towards Nutritional Security and Sustainability. 2021 (pp. 167–180). Springer, Singapore.

  10. Sabreena NM, Mahajan R, Hashim MJ, Iqbal J, Alyemeni MN, Ganai BA, Zargar SM. Deciphering allelic variability and population structure in buckwheat: An analogy between the efficiency of ISSR and SSR markers. Saudi J Biol Sci. 2021;28:6050–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Peng M, Gao Y, Chen W, Wang W, Shen S, Shi J, Wang C, et al. Evolutionarily distinct BAHD N-acyl transferases are responsible for natural variation of aromatic amine conjugates in rice. Plant Cell. 2017;28:1533–50.

    Google Scholar 

  12. Tieman D, Zhu G, Resende MF Jr, Lin T, Nguyen C, Bies D, Rambla JL, et al. A chemical genetic roadmap to improved tomato flavor. Science. 2017;355:391–4.

    Article  CAS  PubMed  Google Scholar 

  13. Kumar V, Singh A, Mithra SVA, Krishnamurthy SL, Parida SW, Jain S, et al. Genome-wide association mapping of salinity tolerance in rice (Oryza sativa). DNA Res. 2015;22:133–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Chen J, Hu X, Shi T, Yin H, Sun D, Hao Y, Chen W. Metabolite-based genome-wide association study enables dissection of the flavonoid decoration pathway of wheat kernels. Plant Biotechnol J. 2020;18(8):1722–35.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Zeng X, Yuan H, Dong X, Peng M, Jing X, Xu Q, Nyima T. Genome-wide dissection of co-selected UV-B responsive pathways in the UV-B adaptation of qingke. Mol Plant. 2020;13(1):112–27.

    Article  CAS  PubMed  Google Scholar 

  16. He Q, Ma D, Li W, Xing L, Zhang H, Wang Y, Du C, Li X, Jia Z, Li X, Liu J. High-quality Fagopyrum esculentum genome provides insights into the flavonoid accumulation among different tissues and self-incompatibility. J Integr Plant Biol. 2023;00:1–19.

    Google Scholar 

  17. Fernie AR, Tohge T. The genetics of plant metabolism. Annu Rev Genet. 2017;51:287–310.

    Article  CAS  PubMed  Google Scholar 

  18. Chen J, Wang J, Chen W, Sun W, Peng M, Yuan Z, Shen S, et al. Metabolome analysis of multi-connected biparental chromosome segment substitution line populations. Plant Physiol. 2018;178:612–25.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Fang C, Luo J. Metabolic GWAS-based dissection of genetic bases underlying the diversity of plant metabolism. Plant J. 2019;97:91–100.

    Article  CAS  PubMed  Google Scholar 

  20. Zhu GT, Wang SC, Huang ZJ, Zhang SB, Liao QG, Zhang CZ, Lin T, et al. Rewiring of the fruit metabolome in tomato breeding. Cell. 2018;172:249–61.

    Article  CAS  PubMed  Google Scholar 

  21. Huang AC, Jiang T, Liu YX, Bai YC, Reed J, Qu B, Osbourn A. A specialized metabolic network selectively modulates Arabidopsis root microbiota. Science. 2019;364(6440):eaau6389.

    Article  CAS  PubMed  Google Scholar 

  22. Martin C, Li J. Medicine is not health care, food is health care: plant metabolic engineering, diet and human health. New Phytol. 2017;216(3):699–719.

    Article  CAS  PubMed  Google Scholar 

  23. Rai A, Saito K, Yamazaki M. Integrated omics analysis of specialized metabolism in medicinal plants. Plant J. 2017;90:764–87.

    Article  CAS  PubMed  Google Scholar 

  24. Keurentjes JJB, Fu J, De Vos CH, Lommen A, Hall RD, et al. The genetics of plant metabolism. Nature Genet. 2006;38(7):842–9.

    Article  CAS  PubMed  Google Scholar 

  25. Schauer N, Semel Y, Roessner U, Gur A, Balbo I, Carrari F, Fernie AR. Comprehensive metabolic profiling and phenotyping of interspecific introgression lines for tomato improvement. Nat Biotechnol. 2006;24(4):447–54.

    Article  CAS  PubMed  Google Scholar 

  26. Joseph B, Lau L, Kliebenstein DJ. Quantitative variation in responses to root spatial constraint within Arabidopsis thaliana. Plant Cell. 2015;27(8):2227–43.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Yang W, Feng H, Zhang X, Zhang J, Doonan JH, Batchelor WD, Yan J. Crop phenomics and high-throughput phenotyping: past decades, current challenges, and future perspectives. Mol Plant. 2020;13(2):187–214.

    Article  CAS  PubMed  Google Scholar 

  28. Zhao H, He Y, Zhang K, Li S, Chen Y, He M, He F, Gao B, Yang D, Fan Y, Zhu X. Rewiring of the seed metabolome during Tartary buckwheat domestication. Plant Biotechnol J. 2023;21(1):150–64.

    Article  CAS  PubMed  Google Scholar 

  29. Norton GJ, Douglas A, Lahner B, Yakubova E, Guerinot ML, Pinson SR, Price AH. Genome wide association mapping of grain arsenic, copper, molybdenum and zinc in rice (Oryza sativa L.) grown at four international field sites. PLoS ONE. 2014;9(2):e89685.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Borovaya SA, Klykov AG. Some aspects of flavonoid biosynthesis and accumulation in buckwheat plants. Plant Biotechnol Rep. 2020;14(2):213–25.

    Article  Google Scholar 

  31. Pang Z, Chong J, Zhou G, de Lima Morais DA, Chang L, Barrette M, Xia J. Metabo Analyst 5.0:narrowing the gap between raw spectra and functional insights. Nucleic Acids Res. 2021;49(1):W388–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, Mitchell SE. A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS ONE. 2011;6(5):e19379.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Heng Li, Durbin R. "Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;5:1754–60.

    Google Scholar 

  34. Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, et al. A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly. 2012;6(2):80–92.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multi locus genotype data. Genet. 2000;155:945–59.

    Article  CAS  Google Scholar 

  36. Earl DA, vonholdt BM. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv Genet Resour. 2012;4:359–61.

    Article  Google Scholar 

  37. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. MEGA5:molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 2011;28:2731–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Sham PC. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Human Genet. 2007;81(3):559–75.

    Article  CAS  Google Scholar 

  39. Lipka AE, Tian F, Wang Q, Peiffer J, Li M, Bradbury PJ, Zhang Z. GAPIT: genome association and prediction integrated tool. Bioinformatics. 2012;28(18):2397–9.

    Article  CAS  PubMed  Google Scholar 

  40. Bradbury PJ, Zhang Z, Kroon DE, Casstevens TM, Ramdoss Y, Buckler ES. TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics. 2007;23(19):2633–5.

    Article  CAS  PubMed  Google Scholar 

  41. Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B, Altshuler D. The structure of haplotype blocks in the human genome. Science. 2002;296(5576):2225–9.

    Article  CAS  PubMed  Google Scholar 

  42. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics. 2005;21:263–5.

    Article  CAS  PubMed  Google Scholar 

  43. Johnson M, Zaretskaya I, Raytselis Y, Merezhuk Y, McGinnis S, Madden TL. NCBI BLAST: a better web interface. Nucleic Acids Res. 2008;36(2):W5–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

SMZ and AM acknowledge the European Commission for funding Erasmus Mobility Programme (2018–2020 and 2020–2023) that facilitated in undertaking part of this research. The authors wish to thank Ganesh Kumar Agrawal and Randeep Rakwal for initiating this collaboration. SMZ also acknowledges Hon’ble Vice Chancellor, SKUAST-Kashmir (Prof. Nazir A. Ganai) for his support and encouragement for the international collaboration. SMZ also acknowledges DBT,New Delhi for sanction of project on Buckwheat genetics (No. BT/PR45195/NER/95/1930/2022).

Funding

The corresponding author acknowledges the funds received as research grant from NMHS GBPNIHESD, Almora, Uttrakhand, India (Vide Project Sanction Order No. GBPNI/NMHS17-18/SG24/622).

Author information

Authors and Affiliations

  1. Proteomics Laboratory, Division of Plant Biotechnology, Sher-E-Kashmir University of Agricultural Sciences and Technology of Kashmir, Shalimar, Srinagar, Jammu and Kashmir, India

    Sajad Majeed Zargar, Madhiya Manzoor, Amir Bashir Wani & Jebi Sudan

  2. Division of Animal Biotechnology, Sher-E-Kashmir University of Agricultural Sciences and Technology of Kashmir, Srinagar, India

    Basharat Bhat

  3. Division of Genetics and Plant Breeding, Sher-E-Kashmir University of Agricultural Sciences and Technology of Kashmir, Srinagar, India

    Parvaze Ahmad Sofi

  4. Department of Agronomy, Food, Natural Resources, Animals, and Environment, University of Padova, Padua, Italy

    Leonard Barnabas Ebinezer & Antonio Masi

  5. Department of Pharmaceutical and Pharmacological Sciences, University of Padova, Padua, Italy

    Stefano Dall’Acqua

  6. Department of Molecular and Translational Medicine (DMMT), University of Brescia, Brescia, Italy

    Gregorio Peron

Authors
  1. Sajad Majeed Zargar
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Madhiya Manzoor
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Basharat Bhat
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Amir Bashir Wani
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Parvaze Ahmad Sofi
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Jebi Sudan
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Leonard Barnabas Ebinezer
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Stefano Dall’Acqua
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Gregorio Peron
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Antonio Masi
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

SMZ conceived the funding; SMZ and AM designed the experiment; MM, SD and ABW performed laboratory experiments; BB, GP and LBE performed data analysis; MM, BB, SMZ, JS and GP wrote the manuscript with inputs from PAS, LBE and AM. All authors have read and approved the content of the manuscript.

Corresponding authors

Correspondence to Sajad Majeed Zargar or Antonio Masi.

Ethics declarations

Ethics approval and consent to participate

The collection and field evaluation of buckwheat germplasm comply with the institutional guidelines.

Consent for publication

Not Applicable.

Competing interests

Authors declared that they have no competing interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table ST1.

Qualitative results obtained from the HPLC–DAD-MS analysis of Buckwheat samples. Table ST2. Summary of number of effects by type and region. Table ST 3. List of buckwheat genotypes isolated from India and used in present study.

Additional file 2: SF 1.

Chromatogram obtained from the HPLC–DAD analysis of Buckwheat samples. SF 2. Pair-wise alignment between common (y-axis) and Tatarian buckwheat (x-axis). SF 3. LD Plot across the 8 buckwheat chromosomes. SF 4. Haplotype blocks of 8 buckwheat chromosomes. SF 5. Simple illustration of associated biological process of identified genes. SF 6. Simple illustration of molecular functions of identified genes. SF 7. Map showing different collection locations of buckwheat germplasm. The images were obtained from goggle map version 2.1 and are available at https://www.google.co.in/maps/@34.1508271,74.8857874,15z?hl=en&authuser=0.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zargar, S.M., Manzoor, M., Bhat, B. et al. Metabolic-GWAS provides insights into genetic architecture of seed metabolome in buckwheat. BMC Plant Biol 23, 373 (2023). https://doi.org/10.1186/s12870-023-04381-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12870-023-04381-x

Keywords

BMC Plant Biology

ISSN: 1471-2229

Contact us