Introduction

Wheat species are classified based on the hardness of their kernels, with durum wheat having the hardest grains among them. The Latin word durum refers to ‘hard grain wheat’1. Canada is the leading country in durum wheat production, contributing approximately 5.75 million tons2. While production data for Pakistan is very low and is not statistically available at the national level. Durum wheat has strong adaptability and potential in semi-arid regions of the world. Although its global demand is increasing due to high consumption, the production of durum wheat in many countries including Pakistan, which also lies in a semi-arid region remains negligible3.

Durum wheat kernels are characterized as vitreous, amber-colored, hard, and rich in protein, with a naturally nutty taste4. It has great potential to produce semolina due to its unique gluten properties5. Moreover, its grain has suitable biochemical characteristics for manufacturing pasta, noodles, sheeted products, bulgur, breads and durum pastry6.

High quality kernel of durum is due to its carotenoid’s contents in the outermost layer. These naturally occurring carotenoid pigments in durum wheat are divided into xanthophylls, carotenes and unsaturated hydrocarbons. Antioxidant compounds such as β-carotene reduce oxidative damage and enhance the nutritional value of pasta. β-carotene has a role in the inhibition of activity of LOX (lipoxygenase), this automatically enhances quality of pasta7 .

Protein is another important quality component of durum wheat kernel, which influences cooking and nutritional value of bakery products8. Durum wheat contains strong and high gluten content which provides necessary cooking characteristics for good quality pasta production9. The quantity of protein and the quality of gluten in durum wheat is very important for the al dente (an Italian term that means “to the tooth” and refers to the ideal consistency for cooked pasta and rice) parameter of pasta production process10.

The carotenoid pigment and some enzymes like lipoxygenase polyphenol oxidase and peroxidase contribute to the color development and discoloration reactions in pasta. The manufacturing of high-quality pasta requires high pigment content and lower activity of oxidative enzyme11. Hard grain with larger and vitreous endosperm dictated the high yield of semolina12. The translucency or vitreousness of durum wheat grain gives good yield of semolina, reduces the issues like white specks, poor quality and weak strands13. Parameters use to evaluate pasta include wholesomeness, characteristic color, low stickiness & ideal consistency trait and are associated with the durum kernel characteristics14. Lutein and other xanthophylls are the major carotenoid pigments in durum wheat that impart the characteristic yellow color to semolina and pasta15. The presence of starch granules in durum wheat lessens the loss of solids in water during cooking process and also stickiness of pasta surface16.

Pakistani industries rely on imported semolina for pasta production because durum wheat is not widely cultivated in the country, largely due to a lack of awareness. However, the local climatic conditions and soil fertility are well-suited for durum wheat cultivation. Therefore, the objective of this research is to characterize Pakistani durum wheat germplasm based on its nutritional profile and to evaluate its potential for producing high-quality pasta.

Materials and methods

Procurement of raw material

The 11 durum wheat germplasm was taken from the Institute of Plant Breeding and Biotechnology (IPBB) that was sown in the field conditions at the research area of MNS-University of Agriculture Multan during winter season of the year 2018–19. November was mild, while December and January were the coldest months, averaging around 4.5 °C at night. March marked the start of the spring season, with temperatures increasing and a higher chance of light showers from western disturbances. Randomized complete block design (RCBD) was followed with three replicates. Each genotype was sown in 6 m long 5 rowed 6-m2 plot using 50 gram seed in one plot using tractor driven Khan 1 machine. Plots were spaced 1 m head-to-head in 1/2 m apart strips. All recommended agronomic practices were performed during the crop season. The genotypes were harvested during March–April 2019 and used for further analysis. While other raw materials were purchased from local market, Multan. Chemical and reagents used for analysis were purchased from the Duksan, BDH, and Sigma Aldrich, Germany. This whole research was conducted in Central Hi-Tech Laboratories of MNS- University of Agriculture, Multan. The accessions names of these durum germplasm are also provided (Table S1).

Physical characteristics of durum germplasm

A representative sample of selected germplasm was taken, counted manually, mass was determined on weighing balance and represented in grams regarded as thousand kernel weight (TKW)17. The color values of durum wheat grains were measured by Chroma Meter CR-400 (Konica Minolta). To determine color of kernel, the durum wheat grains were placed in a dish and the instrument displayed L*, a* and b* values of color on the screen18.

Preparation of semolina from durum germplasm

Durum germplasms (500 g each) were finely grounded with a micro-whisk mill (Culatti MFC, Switzerland) and passed through a sieve mesh of 250 μm (US Standard 60 mesh) diameter, resulting in 240–270 g semolina for each durum germplasm.

Compositional analysis of durum germplasm

Durum semolina was used to determine major components like fat, ash, protein, fiber and moisture following the procedures explained in19 AACC with method no (30-25, 08-01, 46-19, 32-10 and 44-15 A). The micro components like Fe and Zn were also evaluated by following the principles outlined in AOAC20. Hand wash method no 38 − 10 was used for the estimation of dry and wet gluten content in durum flour (semolina) following the guidelines outlined in AACC19.

Rheological behavior of durum germplasm

The rheological properties of durum flour were measured with mixolab, (Chopin Technologies, Mixolab, France) following method descripted by Vizitiu and Danciu21. The protocol was Chopin S and mixing speed was set at 80 rpm. Mixolab measures dough stability, dough development time and water absorption of the flour just like farinograph values and units.

Manufacturing pasta of durum germplasm

Durum semolina was blended at lowest speed for one minute separately, then addition of water was done in another one minute to make rigid dough in a noodle mixer. The mixing process continued for an additional 8 min to ensure homogeneity, as suggested in standard pasta-making procedures22. The water absorption was analyzed using the Mixolab (Chopin) following21 for all durum germplasms. Water was added to achieve the target dough consistency of 1.1 Nm torque, ensuring precise, genotype-specific hydration rather than a fixed percentage. The dough was then rested for about 10–15 min at a controlled room temperature of 25 ± 1 °C before sheeting, and in dough sheeter, it was sheeted to about 5 mm thickness. The sheeted dough was sliced into proper length suitable for process of drying. In noodle cutter, the sheeted dough was further cut into flat noodles. The prepared samples of pasta (noodles) were dried at 55 °C for 2 h in cabinet tray drier. Prepared samples of every genotype were cooled, packed in foam plate, stored in dark place at room temperature till further analysis.

Physical analysis of pasta of durum germplasm

Firmness of cooked pasta having solid and uniform cross section was measured by texture analyzer (method no 66-50)19. Two beakers were prepared with 300 ml distilled water each, boiled on hot plate, samples of pasta were weighed (0–25 g) and broke into short pieces (almost 5 cm). The samples were cooked in boiling water; after reaching cooking time the sample was drained into Buchner funnel. At consistent temperature, samples were washed with distilled water for 30 s. At room temperature, cooked samples were put into distilled water. The simple strand of pasta was put in the middle of the measuring area of instrument, and it was cut with the help of a knife blade (thickness 1 mm). The value of pasta firmness was recorded as the highest value of force. In the manufacturing of pasta, the loss in cooking and cooked mass was determined by using method described in method (66-50) of AACC19. Pasta sample (20–25 g) was put in boiling distilled water (300 ml). Pasta samples were removed after 30 min interval, washed and drained for 2 min. Loss in cooking was evaluated by given expression:

$${\text{Cooking}}\:{\text{loss}}\:\% = \frac{{{\text{Weight}}\:{\text{of}}\:{\text{drained}}\:{\text{residue}}\:{\text{in}}\:{\text{cooking}}\:{\text{water}}}}{{{\text{Weight}}\:{\text{of}}\:{\text{uncooked}}\:{\text{pasta}}}} \times 100$$

Analysis of color for pasta was done by Chroma meter following the principles (Chroma Meter CR-400 by Konica Minolta) explained in AACC19. Chroma meter gave three parameters of color i.e., L*, a* and b* values. The colorimeter was calibrated by using a white standard plate in such a way that “a values” describe redness (positive) and greenness (negative), “b values” describe the yellowness (positive) and blueness (negative) and “L values” describe black to white (0–100). Colorimeter reads 100% reflectance on blue, red and green scales. The reflective percentage was read on blue and green scales (Z and Y values, respectively). Each sample was put under the lens and reading was noted.

Sensory evaluation of durum germplasm

The sensory traits of manufactured pasta were determined in scores as per the instructions23. The sensory assessment of pasta was done by a panel of 15 members within the age range of 30–40 years consisted of both male and female, selected on the base of their consent, they were trained also. The samples of pasta manufactured from every germplasm were served in white plate (coded) and response was collected utilizing a hedonic scale of 9 points ranged between 1 and 9 with one appointed for minimal score as dislike extremely and 9 greatest score as like extremely. The facility of water was also given to rinse the mouth between evaluations. The coded samples were evaluated by trained panelists as per the instruction for taste, firmness, flavor, color, texture and overall acceptability. The sensory evaluation was done in a quiet, odorless and well-ventilated environment.

Statistical evaluation

All analysis were performed three times and the results were presented as mean ± standard deviation. In the quality attributes of durum wheat germplasm difference was determined using descriptive statistics i.e. mean ± standard deviation. The one way ANOVA (analysis of variance) under CRD (completely randomized design) is used to identify significant difference among attributes of products according to the guidelines explained by Montgomery24. When ANOVA indicated significant differences (p < 0.05), Tukey’s HSD test was applied for post-hoc mean separation. Statistics 8.1 was used to compute analysis of variance and mean comparison whereas mean ± standard deviation was computed through Excel 2013.

Principal component analysis was used to determine the relationship between the studied cases and parameters. Statistica software (version 12.0, StatSoft Inc., Tulsa, OK, USA) was used for statistical analyses. Principal component analysis (PCA), analysis of variance, and correlation was performed at the significance level of α = 0.05. The data matrix used for PCA statistical analysis of the results of physico-chemical characteristics of durum germplasm had 11 rows and 6 columns. The matrix for PCA analysis for compositional analysis of durum germplasm had 11 rows and 8 columns, for rheological characteristics of durum germplasm had 11 rows and 3 columns, and for physical characteristics of paste made from durum germplasm had 12 rows and 6 columns. The optimal number of principal components obtained in the analysis in both cases was determined based on the Cattel criterion. Each input matrix was automatically re-scaled.

Results and discussion

Physical parameters of durum germplasm

Durum germplasms were assessed for physical characteristics i.e. TKM (Thousand Kernel Weight) and color. In accordance with the results achieved, the accessions exhibited suitable kernel size. Data showed that color values varied among different durum accessions. The highest value for TKM (42.98 g) was shown by D-15739 and lowest value was shown by D-16744 (33.14 g) (Table 1). Current work confirms the measures of several publications regarding thousand kernel weight of durum wheat ranged as 32.40–45.66 g25. The grading of wheat grain was dependent on color values and TKM was the potential determinant for milling yield. TKM value has been affected by the specific genotype and environment conditions (including temperature, photoperiod, soil heterogeneity) providing differential growth condition for wheat accessions26,27,28

Table 1 Physico-chemical characteristics of durum germplasm (Values indicate mean ± standard deviation).
Full size table

Mean values of color of durum wheat varieties are given in Table 1, which showed that differences in color values were high in durum wheat accessions. D-15722 had the highest L* color values (L*=86.74), D-16730 (a*=2.69) and D-16738 (b*=30.87), while lowest color values was depicted in D-16744 (L*=62.65), D-15722 (a*=1.47), D-16744 (b*=26.54). Present results resemble with the results of Wrigley et al.29 who published the range of color values of durum wheat grains as 87.4 to 87.8 (l*), 2.8 to 3.0 (a*), 28.7 to 30.6 (b*). Seed color in durum has been controlled through multigenic and number of effective genetic factors ranged between 9 and 2730. Variations can be due to temperature, growing conditions and genetic make-up of durum wheat varieties29.

Compositional analysis of durum germplasm

The results of compositional analysis of durum germplasms showed variations among all parameters. Data related to dry and wet gluten content (Table 1) showed significant (P ≤ 0.05) variation among durum accessions. The highest value of dry and wet gluten contents (9.67% & 28.61%, respectively) was seen in ‘‘D-16730’’, whereas the lowest percentage (8.41% & 24.57%, respectively) was shown in ‘‘D-16744’’. Present results are in agreement with29 who recorded dry gluten level in the range of 9.5–10.3%, while25 determined wet gluten content of durum wheat varieties ranged from 17.29 to 25.87%. The gluten content is directly correlated with protein and was impacted due to circumstances of environment like rainfall, temperature and humidity.

The highest mean values of protein, moisture, crude fat, ash, fiber and NFE were observed in D-15728 (13.31%), D-15722 (12.35%), Durum-97 (2.05%), Durum-97 (1.18%), D-15730 (3.01%), and D-16744 (74.61%), while the lowest mean values were recorded as 10.05% (D-15722), 7.88% (Durum-97), 1.033 (D-15729), 0.59 (D-15722), 1.99 (D-15729), 71.52 (D-15730) respectively (Table 2). There are several factors which have impact on protein content like genetic and non-genetic factors (storage conditions, fertilizer, soil and environment)31. Genotype D-15728 likely exhibits superior gluten and protein quality due to its favorable genetic background, specifically beneficial HMW (high molecular weight) and LMW (low molecular weight) glutenin alleles. These contribute to a stronger network of gluten and an improved glutenin-to-gliadin ratio, resulting in better dough strength and pasta making potential32.

Table 2 Compositional analysis of durum germplasm (Values indicate mean ± standard deviation).
Full size table

The moisture content showed storability of grains of durum and was important consideration for profit in milling. The variations in post-harvest practices and climate may be responsible for changes in moisture percentage33 told a strong genotype effect of environment on ash content variations. Likewise, nitrogen free extract varied among durum germplasm due to the differences in moisture content of durum germplasm. Present findings related to NFE and moisture are in close resemblance with Alamri et al.34 and Amin et al.35 as they noticed high variation in these parameters. The results of proximate composition of durum germplasm are in resemblance with the observations of published literature10,36,37,38,39, ranged as 1.98% to 2.79% (crude fat), 9.5% to 13.5% (crude protein), 0.82–0.89% (ash), 2.49 to 2.58% (crude fiber) and 71.50 to 75.60% (NFE) respectively.

Data regarding mineral (Iron and Zinc) values have been given in Table 2, showing variations among durum germplasm related to these parameters. The Fe content ranged between 19.51 ± 0.08 (D-15722) to 51.08 ± 0.03 (D-15730), similarly, Zn content ranged between 0.41 ± 0.02 (D-16744) to 3.64 ± 0.01 (D-16738). Likewise38, figured iron concentration in durum wheat ranging from 39.19 to 41.88 ppm. Current measures are in anticipation with Rachoń et al.36, they noticed the mean value of zinc content ranging from 53.7 to 54.7 ppm in durum wheat. Variations in iron and zinc content in durum wheat varieties are due to the genetic differences, varietal differences and composition of soil. The color of pasta is influenced by lipoxygenase (LOX) activity and ash content of flour. Higher ash content can darken pasta owing to increased bran and mineral content, while flours with low ash content produce brighter yellow pasta. Germplasms with lower effective LOX activity likely retain more carotenoids, increasing yellowness, whereas higher LOX activity may lead to carotenoid oxidation and paler pasta40.

Rheological behavior of durum germplasm

Rheological parameters of durum flour were estimated through studies of mixograph, and data is represented in Table 3. Graphical representation of mixograph is given in Fig. 1. Variations among durum accessions were observed for parameters such as dough development time (DDT), water absorption (WA) and dough stability (DS). Water absorption value in durum wheat varieties is ranging between 58.6 and 63.6%41. determined water absorption in the range from 53.9 to 72.5% and mean was 64.5% which agrees with the present study. In another study42, we found the water absorption ranging as 53.9 to 67.3% which match with the current study.

Table 3 Rheological characteristics of durum germplasm (Values indicate mean ± standard deviation).
Full size table

Water absorption of durum wheat flour was affected by hardness of durum wheat16 recorded dough development time in the range as 3.5 to 11.0 (min) which is in confirmation with the present findings41 noticed time for dough development ranged as 1.3 to 19.0 (min) and mean was 4.4 (min) which agrees with present research. It is interpreted that variations in dough development time are consequences of differences across durum wheat varieties.

The recorded highest value for dough stability is 22.5 min in D8 and lowest value is 12 min in D3. In past study41, found dough stability range from 1.1 to 27.0. (min) and mean was 6.6 (min) which is in clear agreement with present results. Barakat et al.43, exhibited a range of dough stability as 2.80 to 19.10 (min) which is in conformity with the measures of current work. The present study illustrates that variations in stability of dough are consequences of genetic variations in durum wheat varieties. DDT, WA and DS were in direct correlation to flour quality, recommending that flour with good quality would have greater value for all the above-mentioned characteristics. In accordance with the results, they are suitable for pasta production.

Physical analysis of pasta made from durum germplasm

Pasta made from various durum germplasm was used to estimate firmness using texture analyzer while color was evaluated with the help of Chroma Meter. Samples (25 g) of pasta were used in determining cooking loss and cooked weight of pasta and resultant data ranged between 2.62 and 7.66% and 22.28 to 34.83 g, respectively (Table 4).

Table 4 Quality attributes of pasta made from different durum germplasm.
Full size table

The firmness of pasta ranged between 261.66 and 361.66 g (Table 4), can be attributed to their higher gluten strength and protein (Table 2), which contribute to more elastic and cohesive matrix of dough during pasta manufacturing. That strong network of gluten lessens the leaching of starch during cooking, resulting in lesser cooking loss values (< 8%), the value that was regarded as desirable for high quality pasta preparation from semolina44. The tested semolina samples exhibited relatively consistent performance, suggesting that the selected germplasm of durum possessed comparable protein content and gluten quality, keeping up constant behavior of dough during the whole process16.

The cooked weight of pasta was related with the protein-gluten network (Table 2). A strong network had less residual losses in the cooking water and high cooked weight of pasta45. In present work, the accession D-16732 (34.83 ± 1.07ab) gave the highest value of cooked weight, whilst the highest value of firmness was observed in Durum-97 (361.66 ± 1.53b). The firmness of cooked pasta was influenced with gluten-protein content in durum flour. Accessions with high protein value also had increased firmness of cooked pasta corresponding to Bao et al.45, they revealed that a well-developed network of gluten increases texture of pasta by supplying considerable resistance to deformation in cooking process.

The results of color values of cooked pasta were given in Table 4 suggesting that there was high variation among color values of cooked pasta. The variation in color (L*) of pasta made from durum wheat may be due to bran content and semolina purity as pasta brightness decreases with higher bran contamination in durum wheat varieties. Differences in color (b*) was attributed towards the variation in carotenoid pigment content in durum varieties46. The diversity in observed color values demonstrates both the intrinsic pigment profile of each germplasm line and compositional attributes of durum semolina.

Sensory evaluation of pasta made from durum germplasm

The obtained data related to sensory assessment of pasta (Fig. 2) depicted that pasta made from various durum germplasm have high degree of variation in relation to color, appearance, smoothness, palatability, elasticity, stickiness and overall acceptability. Range of scores related to color, appearance and smoothness were ranged as 5.18 to 6.82, 5.21 to 6.76 and 5.26 to 6.88, respectively. Likewise, the parameters like elasticity and stickiness were ranged from 5.41 to 6.21 and 5 to 6.09, respectively. However, in terms of palatability pasta made from “D-15728” got greatest score (6.59), while pasta prepared from “D-15739” got minimal score (5.41). Similar trend of greatest score was seen in overall acceptability where pasta made from “D-16732” got greatest score, whereas pasta made from “D-15739” got minimal score.

Fig. 1
figure 1

Mixograph curves representing the rheological behaviour of durum wheat varieties.

Full size image
Fig. 2
figure 2

Sensory evaluation scores of pasta prepared from different durum wheat germplasms, showing ratings for color, appearance, smoothness, palatability, elasticity, stickiness, and overall acceptability. Error bars represent ± standard deviation.

Full size image

The color in pasta is affected by the activity of enzyme lipoxygenase in semolina during the processing of pasta and on the amount of ash content in durum grains29. Varietal differences and carotenoid pigment content are main reasons for differences in color score of pasta46. Genetic make-up of durum wheat may be responsible for variations in appearance or look of pasta45. Elasticity of pasta highly depends on the gluten protein content of durum wheat. The elasticity of cooked pasta is in inverse correlation with gluten and protein components in durum germplasm47.

Principal component analysis (PCA)

  1. a.

    Physico-chemical characteristics of Durum germplasm

Conducting the principal component analysis (PCA) allowed us to obtain six new variables that explain the variability of the system by 100%. The PCA analysis shown that the first two principal components (PC1) and PC2 explained the variability of the system by 68.73%. Only the parameters D.G. (%), W.G. (%) and (L*) have a strong influence on the variability of the system because these are contained between the two red circles (Fig. 3A; Table 5). The parameters D.G. (%) with W.G. (%) and (L*) with (b*) are strongly and positively correlated, although (b*) has a weaker influence on the variability of the system than the other three parameters. A strong and negative correlation was observed between the parameters (L*), (b*) and (a*). However, W.G. (%) and D.G. (%) are largely independent, low-level correlations were observed with other parameters.

Fig. 3
figure 3

Projection of variables (parameters on PC1 and PC2 loadings plot) (A), Projection of sample type on PC1 and PC2 scores plot (B).

Full size image
Table 5 Correlation heat map and analysis for the tested samples for the physico-chemical characteristics of durum germplasm.
Full size table

The PCA analysis showed that the first principal component PC1 described the differences between the Durum-97, D-16744, D-15729 and D-15739 varieties and D-15728, D-16730, D-16738, D-16743, D-15730, D-15722 and D16732 varieties in 46.04% (Fig. 3B). Positive values of the principal component PC1 described the results for Durum-97, D-16744, D-15729 and D-15739 varieties, and negative values of the principal component PC1 described the results for D-15728, D-16730, D-16738, D-16743, D-15730, D-15722 and D16732 varieties. PC2 in 22.70% described the differences between Durum-97, D-16744, D-15729, D-15728, D-16730, D-16738, D-16743, varieties D-15739, D-15730, D-15722, and D16732. Positive values of the principal component PC2 described the results for varieties Durum-97, D-16744, D-15729, D-15728, D-16730, D-16738, D-16743, while negative values of the principal component PC2 described the results for varieties D-15739, D-15730, D-15722 and D16732.

The conducted PCA analysis also shown that the parameters D.G. (%), W.G. (%) best described the D-15728, D-16730, D-16738, D-16743 varieties, the parameters (L*) with (b*) efficiently described the D-15730 variety. TKW (g) parameter indicated the best D-15722 and D-16732 varieties, and the parameter (a*) best described Durum-97 and D-16744.

  1. b.

    Compositional analysis of Durum germplasm

Conducting the Principal Component Analysis (PCA) allowed us to obtain eight new variables that explained the variability of the study system by 100%. The PCA analysis showed that PC1 and PC2 explained the variability of the system by 67.04%. The parameters Ash (%), NFE (%), Moisture (%), Crude Fiber (%), Crude Protein (%) and Fe (ppm) have a strong influence on the variability of the system because. These were contained between the red circles (Fig. 4A; Table 6). The Zn (ppm) has a weaker influence, while Crude Fat (%) has a very weak influence. The parameters Crude Fiber (%), Crude Protein (%) and Fe (ppm) are strongly and positively correlated. There is also a positive and strong correlation between Moisture (%) and Zn (ppm), although Zn (ppm) has a weaker influence on the variability of the system. Strongly but negatively correlated parameters are Ash (%), Moisture (%) and Zn (ppm), whereas weakly and negatively correlated are parameters Crude Fiber (%), Crude Protein (%), Fe (ppm), and NFE (%). The weakly and positively correlated parameters are Ash (%) and NFE (%). No correlation was observed between parameters Ash (%) and Crude Fiber (%), Crude Protein (%) and (ppm).

Fig. 4
figure 4

Projection of variables (parameters on PC1 and PC2 loadings plot) (A), Projection of sample type on PC1 and PC2 scores plot (B).

Full size image
Table 6 Correlation heat map of the tested samples for the compositional analysis of durum germplasm.
Full size table

The PCA analysis for compositional analysis of durum germplasm shows that the first principal component PC1 describes the differences between varieties D-16744, D-15729, D-15739, D-15722 and D-16743 in 36.26%, and Durum-97, D-16732, D-16730, D-15728, D-16738 and D15730 (Fig. 4B). Positive values of the principal component PC1 describe the results for D-16744, D-15729, D-15739, D-15722 and D-16743, and negative values of the principal component PC1 described the results for Durum-97, D-16732, D-16730, D-15728, D-16738 and D15730. PC2 in 35.43% mentioned the differences between Durum-97, D-16732, D-16730, D-16744, D-15729, D-15739, and D-15722, D-16743, D-15728, D-16738 and D15730. Positive values of PC2 described the results for Durum-97, D-16732, D-16730, D-16744, D-15729, D-15739, while negative values of PC2 described D-15722, D-16743, D-15728, D-16738, and D15730.

The conducted PCA analysis also showed that Ash (%) parameter entitled the best Durum-97 variety. The NFE (%) parameter described best in D-16744, Moisture (%) in D-15722, Zn (ppm) in D-16743, and Crude Fiber (%), Crude Protein (%) and Fe (ppm) parameters in D-15728 and D-15730.

  1. c.

    Rheological characteristics of Durum germplasm

Conducting the Principal Component Analysis (PCA) allowed us to obtain three new variables that explain the variability of the system by 100%. The PC1 and PC2 explained the variability of the system (82.98%). All parameters have a strong influence on the variability of the system because these are present in between two red circles (Fig. 5A; Table 7). The parameters DDT (min) are weakly and positively correlated with DS (min). A negative and weak correlation occurred between DDT (min), and DS (min), and WA (%).

Fig. 5
figure 5

Projection of variables (parameters on PC1 and PC2 loadings plot) (A), Projection of sample type on PC1 and PC2 scores plot (B).

Full size image
Table 7 Correlation heat map and analysis for the tested samples for the rheological characteristics of durum germplasm.
Full size table

The PCA analysis shown that PC1 described the differences between Durum-97, D-16738, D-15728, D-16730, D-16732, D-16744 varieties and D-16743, D-15730, D-15722, D-15729 and D-15739 varieties in 55.89% (Fig. 5B). Positive values of PC1 results for the D-16743, D-15730, D-15722, D-15729 and D-15739 varieties, while negative values of PC1 described the results for Durum-97, D-16738, D-15728, D-16730, D-16732, and D-16744 varieties. The PC2 in 27.09% described the differences between D-15728, D-16730, D16732 D-16744, D-15730, D-15729 and Durum-97, D-16738, D-15722, D-15739, D-16743 varieties. Positive values of PC2 described the results for D-15728, D-16730, D16732, D-16744, D-15730, D-15729, and negative values of PC2 described the results for Durum-97, D-16738, D-15722, D-15739, and D-16743.

The conducted PCA analysis showed that WA (%) was best in D-16743, D-15730, D-15722, D-15729, and D-15739 varieties, DS (min) in D-16738 variety, and DDT (min) in D-15728, D-16730, D-16732, and D-16744.

  1. d.

    Physical characteristics of pasta made from Durum germplasm

The Principal Component Analysis (PCA) allowed us to obtain six new variables that explained 100% variability of the system. The components PC1 and PC2 explained the 72.15%. variability of the system. All parameters except (b*) have a strong influence on the variability contained between the red circles (Fig. 6A; Table 8). The Firmness (g) and (L*) was strong and positively correlated. There is also a positive correlation between Cooked weight (g) and Cooking loss (%). The parameters (a*) and Cooked weight (g) and Cooking loss (%) were strongly but negatively correlated. There was no correlation between the Firmness (g), (L*), and remaining parameters.

Fig. 6
figure 6

Projection of variables (parameters on PC1 and PC2 loadings plot) (A), Projection of sample type on PC1 and PC2 scores plot (B).

Full size image
Table 8 Correlation heat map and analysis for the tested samples for the physical characteristics of pasta made from durum germplasm.
Full size table

The PCA analysis for physical characteristics of paste made from durum germplasm shown the differences between Control, D-16730, D-16743, D-16744, D-16738, D-16732 and Durum-97 varieties (43.03%), and D-15728, D-15739, D-15722 and D15730 varieties (Fig. 6B). Positive values of PC1 results of Control, D-16730, D-16743, D-16744, D-16738, D-16732 and Durum-97, and negative results for D-15728, D-15739, D-15722 and D15730. The PC2 in 29.10% described the differences between Control, D-16730, D-16743, D-15728, and D-15729, D-16732, Durum-97, D-15722, and D15730. Positive values of PC2 results for Control, D-16730, D-16743, D-15728, and negative PC2 described D-15729, D-16732, Durum-97, D-15722, and D-15730.

The PCA analysis explained the best parameters (a*), Firmness (g) and (L*) in Control variety, the Cooked weight (g) and Cooking loss (%) parameters describe D-16732 and Durum-97 best.

Correlation matrix heat maps

The correlation matrix is a table showing the correlation coefficient for all possible combinations of pairs of variables (Tables 5, 6, 7 and 8). The table’s row and column headers contained the names of variables and inside the numerical values of calculated correlation coefficients. The value of correlation coefficient is in closed range of − 1, 1. The larger its absolute value, the stronger the linear relationship between the variables. The sign of the correlation coefficient indicated a positive or negative correlation between the studied variables. A heat map is a graphical presentation of data in which the individual values contained in the matrix are represented by a color scale. The heat map visualizes data through color differences, which can be seen on the chart as differences where there are higher values.

Conclusion

Durum germplasm showed high variation in TKW, color, proximate, mineral and rheological parameters. Thousand kernel weight is the most important in determining the overall yield of semolina in durum germplasm. It indicates that if TKW is more than yield of semolina will also be greater. Protein showed a strong positive correlation with firmness of cooked pasta suggesting that higher content of protein in durum germplasm leads to greater value of firmness in cooked pasta. Pasta elasticity is dependent on gluten-protein content in durum germplasm indicating that high protein content in durum germplasm gives highest elasticity in cooked pasta. The color of pasta is highly dependent on the ash content in durum germplasm and lipoxygenase activity in durum. It narrated that pasta made from “D-15728” disclosed best results in rheological characteristics and sensory assessment. Likewise, genotype D-15728, characterized by a protein content of 13.31%, exhibited a cooking loss of only 2.87% and a firmness of 292.66 g. On the basis of the discussion parameters D-15728 should be grown for the manufacturing of pasta with best quality parameters.