In quantitative analysis, data were recorded for nine quantitative biochemical parameters, including thiosulfinate content (µmol/g fresh weight (FW)), antioxidant capacity measured as DPPH AE (µmol/g FW) and FRAB (mg/ml), total flavonoid content (TFC) expressed in QE (mg/g FW), total phenolic content (TPC) in GE (mg/g FW), total pyruvic acid (TPA), basal pyruvic acid (BPA), enzymatically produced pyruvic acid (EPA)(µmol/gm FW), total sugar content (g/100g FW), and reducing sugar content (g/100g FW). Each parameter was analyzed in triplicate to ensure accuracy and reliability. Each parameter was analyzed separately, with three independent biological replicates performed for statistical robustness. The detailed methodology for each parameter is provided below.

All chemicals used for biochemical analysis were of analytical grade and procured from Thermo Fisher Scientific, Mumbai, India. Additionally, high-purity analytical reagents were sourced from Sigma-Aldrich, Mumbai, India, ensuring precision and reliability in experimental procedures.

Means and standard errors were calculated for all variables for each species using SAS 9.3, GLM Proc. In addition, cluster analysis, principal component analysis and correlation analysis of the quantitative biochemical traits of the 19 Allium germplasm was estimated by considering a bootstrap analysis with three replications and the tree was constructed by using hierarchical clustering based on Ward (1963) method by JMP 10.3 Pro (by SAS 9.3 base). The correlation graph was constructed using R studio software with package (library (Hmisc) specifically rcorr (as.matrix (mtcars [, c (“mpg”, “cyl”, “disp”, “hp”)]))).

The genus Allium (onion, garlic, chives, and related species) is rich in bioactive compounds such as organosulfur metabolites, phenolics, flavonoids, saponins, and sterols, which contribute to antioxidant, antimicrobial, anti-inflammatory, and other health-promoting activities, making them valuable for bioprospecting. Qualitative phytochemical screening provides a quick and cost-effective means to detect these metabolites, with solvent choice (aqueous vs. hydroalcoholic) influencing recovery of polar and semi-polar compounds.

In results, a qualitative phytochemical analysis was conducted on 19 Allium germplasm representing 15 Allium species. The presence (Y) or absence (X) of key biochemical compounds was assessed across multiple biochemical tests. The phytochemical analysis of various Allium species using distilled water and 80% methanol as a solvent revealed the presence of several polar and semi-polar bioactive compounds ( Supplementary Tables S1 , S2 ). Carbohydrates were found to be consistently present across all tested species, indicating their fundamental role in these plants. Proteins were not strongly detected by the qualitative assays (NaOH-CuSO₄ and Millon’s tests), suggesting their occurrence in relatively low concentrations or limitations of these colorimetric tests in Allium extracts. Similarly, starch was not detected in any of the species.

Secondary metabolites such as phenols, flavonoids, and flavonols were present in all species, highlighting their potential antioxidant properties. Additionally, saponins were found in every species tested using ethanol solvent, suggesting their possible role in plant defense mechanisms and medicinal properties ( Supplementary Table S2 ). Interestingly, tannins were absent in all samples, implying that these species may have a lower astringency compared to other plants rich in tannins. The screening for alkaloids, using Mayer’s, Dragendorff’s, Wagner’s, and Hager’s tests, yielded negative results for all species, indicating the presence of these nitrogenous compounds in trace amount. Similarly, glycosides were mostly absent, except for a positive result in the Salkowski test, suggesting the presence of steroidal or terpenoid glycosides in all species.

Furthermore, the presence of amino acids across all species suggests that these Allium plants could serve as a source of essential nutrients. Fats were not specifically analyzed in this study. Overall, this phytochemical investigation confirms that Allium species are rich in carbohydrates, phenols, flavonoids, saponins, glycosides (as indicated by the Salkowski test), and amino acids, while proteins, starch, tannins, and alkaloids were present in trace amount. These findings contribute to the understanding of the medicinal and nutritional potential of Allium species, supporting their use in traditional and modern bioprospecting applications.

Standard assays enable broad profiling of key constituents, including characteristic sulfur-rich compounds, thereby facilitating the assessment of phytochemical diversity in Allium germplasm. The descriptive statistics for twelve biochemical traits in the evaluated Allium germplasm revealed pronounced phenotypic variability ( Table 2 ). Mean values spanned from 0.50 µmol g⁻¹ FW for enzymatically produced pyruvic acid to 36.70% for carbohydrate content. Among phytochemicals, total flavonoid content (23.06 mg 100 g⁻¹ FW) and total phenol content (22.22 mg 100 g⁻¹ FW) were most abundant, exhibiting wide ranges of 10.42–48.42 mg 100 g⁻¹ FW and 14.35–38.47 mg 100 g⁻¹ FW, respectively. The highest coefficients of variation (CV) were observed for enzymatically produced pyruvic acid (35.72%), total antioxidant capacity (FRAP assay) (29.67%), and basal pyruvic acid (28.26%), indicating substantial metabolic heterogeneity. Variance was greatest for total flavonoid content (143.66), whereas carbohydrate content showed the highest absolute mean with comparatively low variability (CV = 16.03%). Reducing sugar content ranged from 0.51 to 8.79 g 100 g⁻¹ FW (CV = 25.48%), while protein content varied between 0.36 and 1.99 mg ml⁻¹ (CV = 20.68%). These results highlight considerable biochemical diversity, likely reflecting both genotypic differences and environmental influences, and identify key traits of interest for genetic improvement and bioprospecting ( Table 3 ).

Principal component analysis (PCA) revealed that biochemical variation in Allium germplasm is structured across seven principal components (Prin1–Prin7), with the first three components explaining the majority of the variability ( Table 4 ; Figure 4 ). PC1 (Prin1) showed high positive loadings for total flavonoid content (0.8613), antioxidant activity by FRAP (0.5598), total phenol content (0.6676), total pyruvic acid (0.7035), basal pyruvic acid (0.7194), and carbohydrate percentage (0.7574). This axis represents a “phenolic–pyruvic–carbohydrate complex”, indicating genotypes with higher values for these traits also tend to have elevated antioxidant potential and flavor-related compounds. PC2 (Prin2) was dominated by strong positive loadings for total sugar (0.8631), reducing sugar (0.6624), allicin/total thiosulfinate content (0.7166), and basal pyruvic acid (0.5544). This suggests a “sugar–thiosulfinate association”, potentially linked to both metabolic energy content and flavor intensity. PC3 (Prin3) was heavily influenced by enzymatically produced pyruvic acid (0.7857) and total pyruvic acid (0.5796), while showing negative association with FRAP antioxidant activity (–0.4974) and reducing sugar (–0.4081). This reflects a “pungency–antioxidant trade-off” dimension, where certain pungency-related compounds increase as some antioxidant metrics decline. PC4 (Prin4) showed moderate loadings for allicin/total thiosulfinate content (0.4924) and protein content (0.3519), but negative associations with carbohydrate percentage (–0.4017) and reducing sugar (–0.4918), representing a “protein–thiosulfinate vs. carbohydrate–sugar” contrast. PC5–PC7 accounted for smaller portions of variability, capturing subtler patterns, such as protein–carbohydrate relationships (PC5) and minor variations in antioxidant measures (PC6, PC7) ( Table 5 ; Supplementary Table S5 ).

Overall, PCA separated the traits into distinct biochemical clusters (i) flavonoid–phenolic–pyruvic–carbohydrate group, (ii) sugar–thiosulfinate group, and (iii) enzymatic pyruvate-driven pungency group highlighting biochemical pathways that may be co-regulated or show metabolic trade-offs. These components can serve as biochemical fingerprints for selecting genotypes with targeted nutritional or flavor profiles.

In summary, the evaluation of 19 Allium accessions representing 15 species revealed clear and contrasting patterns of biochemical diversity between cultivated and wild taxa. Cultivated species such as A. cepa, A. sativum, A. fistulosum, and A. cepa var. aggregatum consistently clustered together, characterized by elevated carbohydrate and sugar contents, a reflection of selective domestication for yield, storability, and consumer preference. In contrast, several wild and underutilized species, notably A. hookeri and A. macranthum, were enriched in allicin, pyruvic acid, flavonoids, phenolics, and antioxidant activity, emphasizing their importance as reservoirs of nutraceutical compounds and stress-adaptive traits. Similarly, A. tuberosum and A. chinense displayed distinct organosulfur and antioxidant profiles, while A. schoenoprasum and A. przewalskianum exhibited elevated antioxidant potential and overall phytochemical richness. Such diversity highlights the evolutionary trade-offs between domestication, which favored palatability and carbohydrate accumulation, and wild relatives, which retained greater phytochemical complexity and adaptive metabolites. Collectively, these findings establish wild Allium taxa as valuable genetic and biochemical resources with significant potential for bioprospecting and for breeding onion and garlic cultivars with enhanced nutritional quality, storability, and resilience to biotic and abiotic stresses.

Phytochemical analysis is a fundamental approach in plant science to identify and characterize bioactive compounds. It reveals the chemical composition of plants, providing insights into metabolic diversity and functional properties. It helps to detect valuable metabolites guides the selection of plant species for further pharmacological studies or natural product development. Germplasm screening for bioactive compounds aids in distinguishing chemotypes, identifying superior accessions, and supporting breeding programs. The phytochemical analysis of various Allium species revealed the presence of key bioactive compounds, which align with previous studies on the genus. The phytochemical screening of 19 Allium germplasm representing 15 Allium species using aqueous and 80% methanol extracts revealed a conserved metabolic profile across the genus, with notable implications for their chemotaxonomic classification and potential bioactive applications. The universal presence of carbohydrates, phenolic compounds, flavonoids, flavonols, and saponins aligns with the well-documented secondary metabolite composition of Allium species, which are renowned for their antioxidant, antimicrobial, and anti-inflammatory properties (Lanzotti, 2006).

The presence of carbohydrates in all tested species is expected, as Allium plants are known to be rich in polysaccharides such as fructans, which contribute to their nutritional and medicinal value (Seki et al., 2019; Gonzalez-Morales et al., 2017). These compounds have been shown to support gut health and modulate immune responses (Bhat, 2020). The absence of proteins and starch in all tested species is also consistent with previous reports. The absence of proteins and starch may reflect methodological limitations, as methanol extraction is less efficient for polar macromolecules like proteins, which often require aqueous or buffered solvents (Arena et al., 2024; Harborne, 1998). The negative protein tests suggest that the methanolic extract might not effectively extract proteins, or the species naturally have low protein content in their soluble fraction. Starch is generally present in low quantities in Allium plants, as their primary storage carbohydrates are fructans rather than starch (Benitez et al., 2017). Similarly, the lack of tannins contrasts with some reports in wild Allium species (Mendes et al., 2012), suggesting variability in tannin production across environments or genetic lineages. The uniform absence of alkaloids (confirmed by four distinct tests) further distinguishes Allium from alkaloid-rich genera like Solanum or Papaver, reinforcing its unique chemoprofile (Hesse, 2002).

The widespread presence of phenols, flavonoids, and flavonols aligns with numerous studies highlighting the antioxidant, anti-inflammatory, and antimicrobial properties of Allium compounds (Lanzotti, 2006; Benkeblia et al., 2005). These phenolic compounds contribute to the therapeutic potential of Allium species and have been associated with reducing the risk of chronic diseases such as cardiovascular disorders and cancer (Najeebullah et al., 2021; Kothari et al., 2020). Phenolics and flavonoids, in particular, are key contributors to the health-promoting effects of Allium vegetables, such as garlic (A. sativum) and onions (A. cepa), and their ubiquitous occurrence in this study underscores their evolutionary significance as defense molecules (Brewster, 2008; Assefa et al., 2018).

Saponins, which were found in all species, are known for their antimicrobial, immune-modulating, and cholesterol-lowering effects (Martins et al., 2016). The detection of steroidal/triterpenoid glycosides via Salkowski’s test corroborates earlier studies identifying saponins as hallmark metabolites in Allium (Sobolewska et al., 2016). These compounds, such as the immunomodulatory saponins in A. ampeloprasum, are linked to antifungal and cholesterol-lowering activities (Lanzotti, 2006). Their presence further supports the medicinal use of Allium plants in traditional medicine. In contrast, tannins were absent in all tested species, suggesting a lower astringency compared to other medicinal plants that contain high levels of tannins (Arena et al., 2024; Haslam, 1996). This may indicate that Allium species have a different chemical defense mechanism than tannin-rich plants.

The absence of alkaloids, as confirmed by four different tests, is consistent with the general chemical profile of Allium species, which are not known for significant alkaloid content (Benkeblia et al., 2005). Instead, sulfur-containing compounds such as allicin, alliin, and ajoene are the primary bioactive compounds in Allium species, contributing to their antimicrobial and cardiovascular benefits (Corzo-Martínez et al., 2007; Kothari et al., 2020). The detection of glycosides (positive Salkowski test) suggests the presence of steroidal or terpenoid glycosides, which may contribute to the pharmacological effects of Allium species. Glycosides have been linked to various medicinal properties, including cardioprotective and anti-inflammatory activities (Cai et al., 2004). However, the negative results in Keller-Killiani and Borntrager’s tests suggest the absence of cardiac glycosides and anthraquinone glycosides, respectively. However, the negative results for cardiac glycosides (Keller-Killiani) and anthraquinones (Borntrager’s) align with the genus’s limited association with these glycoside subtypes. Finally, the presence of amino acids in all species highlights their potential nutritional benefits, as Allium plants are known to contain essential amino acids beneficial for human health (Seki et al., 2019). The amino acid profile of these species may contribute to their role in metabolism and overall dietary value. The reported presence of amino acids under the “fats” category requires clarification, as amino acids are nitrogenous compounds typically associated with proteins. While methanol may extract free amino acids, their classification here likely reflects methodological overlap in colorimetric assays (e.g., ninhydrin reactivity), as noted in biochemical screening protocols (Kumar and Kumar, 2023).

Under methodological considerations the use of 80% methanol likely enhanced the extraction of medium-polarity compounds like flavonoids and saponins while excluding highly polar (e.g., proteins) or nonpolar (e.g., essential oils) metabolites. This aligns with studies advocating methanol’s efficacy for phenolic extraction in Allium (Shukla et al., 2024; Arena et al., 2024). However, the absence of alkaloids may reflect solvent incompatibility, as alkaloid isolation often requires acidified solvents.

The uniformity in phytochemical profiles across diverse Allium species supports the theory of conserved biosynthetic pathways, possibly driven by ecological pressures such as herbivory or pathogen resistance. For instance, saponins and flavonoids are known deterrents against microbial and insect attacks. These findings reinforce the chemotaxonomic value of flavonoids and saponins in delineating Allium subgenera, as proposed by Fritsch and Friesen (2002).

The statistical analysis of twelve biochemical parameters in the evaluated Allium germplasm revealed marked phenotypic variability, attributable to genetic diversity, environmental influences, and inherent metabolic differences. Coefficient of variation (CV) values indicated traits with the greatest heterogeneity, notably reducing sugar (95.72%) and enzymatically produced pyruvic acid (65.83%), suggesting dynamic post-harvest metabolic changes likely driven by enzymatic activity and carbohydrate–sulfur metabolism interactions. Such variation is consistent with previous findings reporting cultivar- and storage-dependent fluctuations in sugar accumulation and pyruvic acid production (Gonzalez et al., 2017; Benke et al., 2021; Shukla et al., 2024). High mean values of total flavonoid content (23.06 mg 100 g⁻¹ FW) and total phenol content (22.22 mg 100 g⁻¹ FW) highlight the strong antioxidant potential of the germplasm. These secondary metabolites are widely recognized for their roles in plant defense, stress tolerance, and human health benefits, including anti-inflammatory and free radical scavenging properties (Cai et al., 2004). The broad range observed in total flavonoid content (10.42–48.42 mg 100 g⁻¹ FW) and total sugar (0.51–8.79 g 100 g⁻¹ FW) suggests that certain genotypes and/or growth conditions enhance secondary metabolite biosynthesis, in line with studies linking flavonoid accumulation to genotype–environment interactions and light-mediated regulation (Jaakola, 2013).

The highest variance (143.66) recorded for total flavonoid content further reinforces its sensitivity to genetic expression and environmental modulation. Similarly, variability in pyruvic acid content aligns with its established role as a biochemical marker of pungency in Allium species, where sulfur assimilation and alliinase-mediated reactions are key determinants (Gonzalez-Morales et al., 2017).

Overall, the observed biochemical heterogeneity underscores substantial potential for targeted selection and breeding aimed at enhancing phytochemical richness. Integrating metabolomic profiling with genomic and transcriptomic tools could provide mechanistic insights into the regulation of these traits, enabling the development of Allium cultivars with optimized nutritional and functional qualities (Vukovic et al., 2021).

The hierarchical clustering pattern ( Figure 1 ) distinctly separated the evaluated Allium accessions into two major clusters, reflecting strong underlying biochemical divergence. This grouping was primarily influenced by variability in allicin content, pyruvic acid levels (total, basal, and enzymatic), phenolic and flavonoid concentrations, antioxidant capacity (DPPH and FRAP assays), and sugar composition. Such metabolic differentiation aligns with genetic variation, ecological adaptation, and domestication history within the genus (Kamenetsky and Rabinowitch, 2001; Benke et al., 2024).

Cluster I (red branch in Figure 1 ) predominantly encompassed cultivated bulb-type A. cepa forms (common onion, aggregatum) and A. sativum (garlic). These accessions were characterized by lower concentrations of allicin, phenolics, and flavonoids, but higher carbohydrate and sugar contents. This pattern reflects domestication-driven selection for milder taste, higher storability, and consumer-preferred sweetness, often at the cost of reduced secondary metabolite abundance (Benkeblia and Lanzotti, 2007). Cluster II (green branch in Figure 1 ) comprised a diverse set of wild and semi-cultivated taxa including A. fistulosum, A. hookeri, A. ampeloprasum, A. schoenoprasum, A. chinense, and A. tuberosum characterized by elevated allicin, pyruvic acid, total phenol, and flavonoid levels, coupled with higher antioxidant activity. These phytochemically rich species align with prior observations (Gonzalez-Morales et al., 2017; Kothari et al., 2020; Shukla et al., 2024) that wild Allium possess superior medicinal potential and stress tolerance, likely due to their high investment in defensive secondary metabolites. Pearson’s correlation analysis supported these clustering patterns. Strong positive correlations between total pyruvic acid and basal pyruvic acid (r = 0.9262), and between allicin content and basal pyruvic acid (r = 0.6671), indicate tight co-regulation of pungency-related sulfur compounds—traits that heavily influenced cluster separation. Similarly, moderate correlations between phenolic and flavonoid content (r = 0.6504) and between carbohydrate percentage and FRAP-based antioxidant capacity (r = 0.5170) reflect overlapping biochemical pathways that contribute to trait co-association within clusters. Negative correlations such as between DPPH antioxidant activity and flavonoid content (r = –0.8302) or between carbohydrate percentage and protein content (r = –0.4913) highlight potential metabolic trade-offs. These trade-offs may partly explain why carbohydrate-rich, low-pungency cultivars dominate Cluster I, while phytochemical-dense, high-pungency accessions group in Cluster II.

Overall, the integration of correlation analysis with hierarchical clustering ( Figure 1 ) reveals that Allium biochemical diversity is structured by strong co-regulation of sulfur metabolism, phenolic pathways, and carbohydrate allocation. This metabolic divergence between domesticated and wild species underscores the breeding potential of wild relatives for enhancing health-promoting phytochemicals, antioxidant capacity, and adaptive resilience in cultivated onions and garlic.

The present study highlights the extensive biochemical diversity within Allium species, as revealed by principal component analysis (PCA), aligning with earlier reports that emphasized variability in secondary metabolites, sulfur compounds, and antioxidants across the genus (Kamenetsky and Rabinowitch, 2001; Kumar and Kumar, 2023). The Kaiser–Meyer–Olkin (KMO) statistic (0.65) and a significant Bartlett’s test confirmed the adequacy of the dataset for multivariate reduction, underscoring the robustness of the PCA results. The first two principal components (PC1 and PC2) explained 59.7% of the total biochemical variance, capturing the major axes differentiating cultivated and wild Allium germplasm. Key contributors included total sugar, reducing sugar, allicin content, total pyruvic acid, phenolic content, and antioxidant activity—traits also emphasized in interspecific diversity studies by Jikah et al. (2024) and Havey (2018). Similar analyses have demonstrated that PCA effectively distinguishes garlic (A. sativum) and onion (A. cepa) accessions on the basis of their biochemical fingerprints (Maccelli et al., 2020). The clustering observed in the dendrogram was consistent with PCA groupings, reinforcing the idea that biochemical composition serves as both a taxonomic marker and a functional discriminator in Allium species (Kamenetsky and Rabinowitch, 2001).

The PCA biplot ( Figure 3 ) revealed three major biochemical clusters. The first, an antioxidant–phenolic–pyruvic module represented along PC1, included traits such as total phenol, flavonoid content, carbohydrate percentage, and pyruvic acid indices. This grouping indicates that phenolic metabolism and pungency traits are co-regulated, consistent with reports that phenolics and sulfur-derived pyruvates share overlapping biosynthetic pathways contributing to both antioxidant potential and flavor intensity (Benke et al., 2024). Pearson’s correlation supported this interpretation, showing strong positive associations between total pyruvic acid and basal pyruvic acid (r = 0.9262) and between allicin and basal pyruvic acid (r = 0.6671). These correlations confirm the biochemical link between pungency-related compounds and sulfur metabolism. Cultivated A. cepa genotypes (e.g., Bhima Kiran, Bhima Super, Bhima Shweta) and A. ascalonicum were positioned positively along this axis, reflecting their higher phenolic and pyruvic acid concentrations, which enhance both health-promoting properties and characteristic onion pungency (Martins et al., 2016).

The second grouping, a sugar–thiosulfinate module aligned with PC2, was dominated by strong loadings from total sugar, reducing sugar, and allicin/total thiosulfinate content. A. sativum accessions clustered here, reflecting their elevated thiosulfinate levels and carbohydrate richness—biochemical features long recognized for contributing to garlic’s medicinal efficacy and storability (Benkeblia and Lanzotti, 2007). Pearson’s correlation further supported this association, as total phenol content was moderately correlated with total flavonoid content (r = 0.6504) and basal pyruvic acid (r = 0.5121), suggesting that phenolic metabolism overlaps with sulfur pathways. The close association between sugars and thiosulfinates highlights coordinated regulation of primary and secondary metabolism, whereby carbohydrate flux supports the biosynthesis of organosulfur compounds (Khar et al., 2022). These biochemical linkages explain why sugars influence both taste and postharvest behavior in Allium crops (Benkeblia and Shiomi, 2006; Rabinowitch and Currah, 2002).

A third cluster, driven by PC3 contributions, reflected a pungency–antioxidant inverse relationship. Enzymatically produced pyruvic acid was negatively associated with FRAP-based antioxidant capacity and reducing sugars. Pearson’s correlations confirmed these antagonisms: reducing sugar was negatively correlated with enzymatic pyruvic acid (r = –0.4452), while FRAP activity was negatively correlated with DPPH scavenging capacity (r = –0.4871). Similarly, DPPH-based antioxidant activity was strongly negatively correlated with total flavonoid content (r = –0.8302) and carbohydrate percentage (r = –0.6868). These results suggest that radical scavenging efficiency in Allium is supported by non-flavonoid antioxidants, particularly organosulfur derivatives, which contribute to pungency and antimicrobial activity but do not always act as direct antioxidants (Benkeblia et al., 2005; Seki et al., 2019). The observed inverse associations underscore evolutionary metabolic balancing: selection for pungency in wild relatives often coincides with enhanced sulfur metabolism but lower sugar- or flavonoid-derived antioxidant activity. Conversely, cultivated types exhibit greater carbohydrate accumulation but relatively diluted phytochemical richness. Wild species such as A. macranthum and A. hookeri occupied the negative quadrants of the PCA space, reflecting unique but lower trait concentrations, likely adaptations to ecological niches (Corzo-Martínez et al., 2007; Kamenetsky and Rabinowitch, 2001).

Overall, the PCA and correlation analyses clearly separate domesticated from wild Allium germplasm. Cultivated onions and shallots were enriched in phenolic and pyruvic acid traits, while garlic accessions showed sugar–thiosulfinate dominance. Wild and underutilized taxa, although lower in trait concentrations, represent reservoirs of alleles for stress resilience, pungency, and medicinal properties (Fritsch and Friesen, 2002). The scree plot confirmed that the first three components capture most of the meaningful biochemical variance, supporting the interpretation of three primary axes: (i) phenolic–pyruvic–carbohydrate complex, (ii) sugar–thiosulfinate association, and (iii) enzymatic pyruvate–antioxidant inverse association. Together, these axes reflect distinct metabolic fingerprints within the Allium genus (Havey, 2018).

In summary, Allium biochemical diversity is structured along metabolic pathways that reveal evolutionary balances between primary (sugars, carbohydrates) and secondary (phenolics, flavonoids, organosulfur compounds) metabolism. Integrating PCA and correlation networks provides a framework for exploiting this diversity in breeding programs balancing flavor, nutritional quality, and health-promoting properties while also guiding the conservation of wild relatives as reservoirs of unique metabolic traits.