The proximate composition of the A. tuberosum leaves was analyzed following the standard food analysis methods described in the Association of Official Analytical Chemists. The moisture was determined using a hot air oven as per AOAC 952.08, 2016 (Cunniff and Washington, 1997), ash content was estimated by heating plant sample in a muffle furnace for about 5-6 h at 500°C (Official Methods of Analysis of AOAC, 2016; Cunniff and Washington, 1997) whereas the acid insoluble ash content was determined by dissolving ash in dilute hydrochloric acid (10% w/w) (Official Methods of Analysis of AOAC, 2016; (Cunniff and Washington, 1997), the liquid was filtered through an ashless filter paper and thoroughly washed with hot water. The filter paper was then ignited in the original dish, cooled and weighed. The crude lipid was extracted from a moisture-free sample with petroleum ether (60-80°C) in a Soxhlet apparatus for about 6-8 h. Estimation of crude fiber content in the plant materials was carried out by treating the fat and moisture-free materials with 1.25% dilute acid and 1.25% alkali followed by washing with water and ignition of the residue. The crude protein was determined using the micro Kjeldahl method (AOAC 992.23, 2016; Sarkar et al., 2020). Total fat was determined according to the acid hydrolysis method (AOAC 948.15, 2016; Cunniff and Washington, 1997), using a Soxhlet extractor at 60°C until constant weight. Dietary fiber was determined according to the enzymatic gravimetric method (AOAC 985.29, 2016; Cunniff and Washington, 1997), digested leaf samples with heat using alpha-amylase, protease, and amyloglucosidase at 60°C, respectively, then added ethanol to samples to precipitate fiber (McCleary, 2023; Prosky et al., 1985).

The total carbohydrate content was estimated as described in the method of Hedge et al. (1962) by calculation having estimated all the other fractions by proximate analysis i.e. % Available carbohydrates = 100 - [% Moisture + % Ash + % Fat +% Protein + % Fiber]. The % organic matter was calculated as per James (2013). The energy content was determined by multiplying the values obtained for protein, fat and available carbohydrate by 4.00, 9.00 and 4.00, respectively and adding up the values. Total sugar, reducing sugar, and non-reducing sugar were calculated as per the Lane and Eynon method (James, 2013) which measures sugar as reducing sugar and total sugar as invert sugar. Non-reducing sugar is determined by subtracting the total reducing sugar from the total sugar and multiplying the remainder with a 0.95 factor.

The analysis of the mineral composition of A. tuberosum leaves was done using a spectrometer as per AOAC 990.08:2023 (Szydłowska and Szłyk, 2003; Piatkowska et al., 2015; Isaac and Johnson, 1985). Plant material was taken in a pre-cleaned and constantly weighed silica crucible and heated in a muffle furnace at 400°C till there was no evolution of smoke. The crucible was cooled at room temperature in a desiccator and carbon-free ash was moistened with concentrated sulphuric acid and heated on a heating mantle till fumes of sulphuric acid ceased to evolve. The crucible with sulphated ash was then heated in a muffle furnace at 600°C till the weight of the content was constant. One gram of sulphated ash obtained above was dissolved in 100 ml of 5% hydrochloric acid (HCl) to obtain the solution ready for determination of mineral elements through atomic absorption spectroscopy (AAS) (AA 800, Perkin-Elmer, Germany). A standard solution for each element was prepared and calibration curves were drawn for each element using AAS (Indrayan et al., 2005). Total nitrogen was estimated using the modified Kjeldahl method as per IS 7219 (Bremner and Mulvaney, 1982; Sáez-Plaza et al., 2013; Sarkar et al., 2014). The total Sulphur content was estimated by inductively coupled plasma-optical emission spectrometry (ICP-OES) (iCAP PRO-X-Duo, Thermo-Scientific, MA, USA) after sample preparation with tetramethylammonium hydroxide (TMAH) (Tavares et al., 2020; Hansen et al., 2013).

The total phenolic content of the samples was quantitatively analyzed using the Folin-Ciocalteu reagent, following the procedure (Blainski et al., 2013). Gallic acid served as the standard for this assay. The absorbance of the reaction mixture was measured at a wavelength of 750 nm using a UV-Vis spectrophotometer (Model no. UV1800, Shimadzu Corporation, Japan). The phenolic content in the samples was determined by comparing the absorbance values with a standard curve prepared using known concentrations of gallic acid (1, 2, 5, 7, 10, 20 and 50 ppm). The results were expressed as milligrams of Gallic Acid Equivalent per gram of dry weight (mg GAE/g DW).

The total flavonoid content was determined using a modified version of the aluminium chloride colorimetric method, with quercetin as the reference standard (Sulastri et al., 2018). In this assay, the sample extracts were reacted with aluminium chloride, and the resulting complex’s absorbance was measured at 510 nm. The flavonoid content was quantified by comparing the absorbance readings against a calibration curve generated with quercetin standards by a UV-Vis spectrophotometer (Model no. UV1800, Shimadzu Corporation, Japan). The results were articulated in terms of milligrams of Quercetin Equivalents per gram of fresh weight (mg QE/g FW).

The antioxidant activity of the sample extracts was evaluated using the DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging assay (Scherer and Godoy, 2009). The procedure involved adding 50 µL of the sample extract to 1.950 mL of a DPPH solution, which was prepared in 95% methanol to achieve an initial absorbance of 0.8 to 0.9 at 515 nm. The mixture was incubated at room temperature in the dark for 90 minutes to allow the reaction to occur. Post incubation, the decrease in absorbance was measured at 515 nm using a spectrophotometer (Model no. UV1800, Shimadzu Corporation, Japan). The antioxidant activity was quantified as the percentage of DPPH radical inhibition, calculated using the following formula:

This percentage indicates the capacity of the sample to scavenge DPPH radicals, thereby reflecting its antioxidant activity

The morphological traits of Allium tuberosum accessions exhibited significant differences across various age windows, particularly in plant height, tiller number, crop spread, leaf width, and stem length (Table 1). The plant height ranged from 21.53 cm to 34.61 cm, depending on the accession and age window, with AT1 showing the greatest height at 270 DAT (34.61 cm). Seasonal factors had a more profound impact on plant height than the age window alone, as indicated by higher values recorded during periods of bright sunshine hours i.e. April to June (age window 90 DAT to 180 DAT and 450 DAT to 540 DAT). The tiller counts also increased significantly with age, starting from approximately 2 tillers at 90 DAT and reaching over 43 tillers per plant at 720 DAT, with AT1 consistently producing the highest number of tillers. In terms of crop spread, AT1 exhibited the widest spread of 18.52 cm at 720 DAT, followed by AT3 and AT2.

The variations in leaf width and stem length were similarly pronounced across different age windows. AT1 had the widest leaves at 720 DAT (8.28 cm), while the longest stem length was also observed at 720 DAT, with AT1 and AT3 recording lengths of 7.81 cm and 7.9 cm, respectively. The leaf count showed moderate differences across accessions, with AT1 consistently producing a higher number of leaves, peaking at 5.17 leaves per plant at 540 DAT. These results suggest that both genetic makeup and the age window strongly influenced the morphology of the accessions, with seasonal effects playing an equally important role in shaping plant structure, especially during windows of bright sunshine.

Tables 2 , 3 show that statistical significance was evident for most morphological traits. The least-square difference (LSD) values from the tables confirmed that the differences in plant height, tiller count, crop spread, and other morphological parameters were indeed significant across accessions and age windows, further reinforcing the impact of age and genetic variation on plant structure. This significance emphasizes the necessity of considering both genetic factors and environmental conditions when evaluating the morphology of A. tuberosum.

The crop growth patterns over the study period, specifically regarding tillering, crop spread, and dry matter content, showed steady increases with age across all accessions. Tillering, which began with an average of around 2.5 tillers at 90 DAT, grew progressively to over 43 tillers by 720 DAT. AT1 consistently led in tiller production across all age windows, although the other accessions (AT2 and AT3) followed closely behind. Similarly, crop spread increased with age, starting at 3.75 cm at 90 DAT and reaching a maximum of 18.52 cm at 720 DAT for AT1, while AT3 and AT2 showed slightly smaller, yet significant, values. Dry matter content was influenced by both crop age and seasonal variations. The highest dry matter yields were recorded during the 90–180 DAT and 450–540 DAT windows, coinciding with periods of bright sunshine and elevated temperatures. AT1 maintained the highest average dry matter content (11.18%), while AT2 and AT3 trailed behind slightly. This seasonal trend demonstrates the strong interplay between environmental conditions and crop physiology, with the brightest sunshine hours contributing to higher dry matter accumulation.

The significant differences between accessions, as shown by the LSD values, underscore the importance of considering both genetic traits and environmental conditions when evaluating the crop growth of A. tuberosum. The growth patterns observed suggest that tillering and dry matter content are not only dependent on age but also heavily influenced by external factors such as sunshine hours and temperature, further validating the role of environmental variables in crop development.

The quality of the harvest was assessed through a range of parameters, including green foliage yield, waste generated, the waste-to-yield ratio, and the stem-to-leaf ratio. Fresh foliage yield increased significantly with crop age across all accessions, with AT1 producing the highest overall yield. In the 630-720 DAT window, AT1 yielded 30.75 tons/ha, whereas AT2 and AT3 produced slightly lower yields of 29.67 tons/ha and 29.21 tons/ha, respectively. Over the two years, AT1 accumulated a total yield of 157.01 tons/ha, significantly higher than the other accessions, illustrating its superior productivity. However, the higher productivity also led to a proportional increase in waste generation, with AT1 producing 6.38 tons/ha of waste in the 540-630 DAT window, while AT2 and AT3 followed closely behind. The waste-to-yield ratio was highest during the 180-270 and 540-630 DAT windows, coinciding with the monsoon season, which negatively impacted foliage quality.

The stem-to-leaf ratio, which served as a critical measure of harvest quality, indicated that a higher ratio corresponded to poorer quality, as the crop is primarily cultivated for its leaves. Moreover, the results suggest that seasonal variations, particularly during monsoon windows, had a profound effect on yield quality. The stem-to-leaf and waste-to-yield ratios were notably higher during these periods, confirming that environmental conditions played a critical role in determining harvest quality. These findings, combined with the significant LSD values, further validate the hypothesis that age, genetics, and seasonality collectively influence both yield and its quality (Figures 2, 3).

The proximate analysis of A.tuberosum All-1587 (AT1) revealed a high moisture content of 88.94% in the fresh leaf cuttings ( Table 4 ). This translates to a low caloric density of 36.13 Kcal per 100g. Interestingly, this moisture content is slightly higher than reported by Pooja (2021), suggesting latent variations based on cultivation practices. Rest, the ash content of 1.47 g per 100g, with a minimal acid-insoluble fraction (0.04 g), indicates the presence of dietary minerals. The low-fat content (0.85 g per 100g) and the lower cholesterol (<0.50 g per 100g) of AT1 are promising, especially considering the commendable ratio of unsaturated to saturated fats. The presence of a notable amount of dietary fiber (2.56 g per 100g) and protein (3.10 g per 100g) in AT1 is important. Both, the protein and dietary fiber content are quite higher than the A. cepa bulbs (1.1 g and 1.7g per 100g respectively) (Bhattacharjee et al., 2013; Mahmood et al., 2021).

The analysis of AT1 reveals a rich mineral profile dominated by essential macro-elements. Potassium (5355.30 mg/kg), followed closely by phosphorous (691.68 mg/kg), calcium (1126.99 mg/kg) and magnesium (569.39 mg/kg). In the macro-elements, a noteworthy aspect is the substantial amount of Sulphur (2484.97 ± 104 mg/kg) present in AT1. This aligns with the characteristic pungent aroma of Alliums, attributed to various volatile Sulphur compounds (Abe et al., 2020). Similarly, as the assimilation of selenium is carried out through the metabolic absorption route of sulfur (Barak and Goldman, 1997) the concentration of Se was found 42.17 mg/kg which is in line with the previous findings (Pilon-Smits et al., 2009). The presence of essential micronutrients like manganese (5.62 mg/kg), zinc (5.34 mg/kg), copper (1.62 mg/kg), and iron (39.74 mg/kg) further strengthen the nutritional value of AT. Additionally, the low sodium content (14.72 mg/kg) was also reported. The B, Cu and Zn content was comparatively lower in concentration than the A. cepa leaves as reported by Ferreira et al. (2022).

The vitamin profile of A tuberosum (AT1) reveals a captivating interplay of water-soluble and fat-soluble vitamins, impacting human health ( Table 4 ). AT stands out for its particularly high content of vitamin B4, also known as choline (264.45 mg/kg) which is approximately four times more than in A. cepa bulb as reported by Pareek et al. (2017). While the content of these B vitamins is not exceptionally high, their combined presence contributes to the overall nutritional value of AT1. Interestingly, pyridoxine (vitamin B6) is present in very low quantities (0.053 ± 0.0045 mg/kg), suggesting that AT may not be a significant dietary source for this particular vitamin. The presence of vitamin C (18.12 mg/100g) highlights AT1’s contribution to immune function and antioxidant activity (Carr and Maggini, 2017). This vitamin C quantity is quite less than previous reports where Żurawik and Jadczak (2008) reported 43.0 - 56.1 mg/100g of L-ascorbic acid in the leaves A. tuberosum. The presence of vitamin E (5.38 mg/100g) further strengthens the antioxidant properties of AT1 (Traber and Atkinson, 2007). Vitamin E helps to protect cells from free radical damage, which has been linked to various chronic diseases (Traber and Atkinson, 2007). While vitamin K content is relatively low (85.00 mcg/100g), it can contribute modestly to blood clotting and bone health (Shearer and Newman, 2008). The limited detection of folates (vitamin B9) and biotin (vitamin B7) suggests that AT1 may not be a primary source of these essential nutrients.

The reported biochemical properties of AT1 shed light on some interesting aspects that can influence its sensory characteristics, preservation, and culinary applications ( Table 4 ). A slightly acidic pH of 6.7 in a 10% AT solution indicated its weakly acidic nature (Sánchez-Rodríguez et al., 2019). Further investigation revealed the presence of organic acids (0.36 g/100g) within AT1, evident from the titrable acidity value. The total soluble solids (TSS) measured at 7.93°Brix signifies the concentration of soluble solids present in AT1 which is lower than some previous reports i.e. 8.6°Brix as reported by Pooja (2021). This TSS content is at par with that of Welsh onion leaves (7.51 ± 0.31°Brix) and lower than A. cepa leaves (5.87 ± 0.18) as reported by Masahiro Yuasa et al. (2022).

The biochemical profile of A. tuberosum, revealed a promising composition with various health-promoting compounds ( Table 4 ). The presence of a significant amount of total phenolics (1.187 mg GAE/g fresh weight basis) suggests strong antioxidant property (Bindon et al., 2010). This is further supported by the high free radical scavenging activity (64.45% RSA DPPH). These findings align with observations in other Allium species for their antioxidant properties (Rahman, 2007). Interestingly, A. tuberosum displayed a unique profile of organosulfur compounds. While allicin, the characteristic pungent compound of garlic (A. sativum), was present at relatively low levels (16.60 mg kg^-1), the dominant organosulfur compound was allyl methyl thiosulfinate (AMThs) at a considerably higher concentration (269.00 mg kg^-1). This suggests A. tuberosum might possess a distinct flavor and potentially different biological activities compared to garlic (Lanzotti, 2006). The presence of flavonoids (3.19 mg/g) and ascorbic acid (18.12 mg/g) adds further value to the nutritional profile of AT1.