3.1 Electrophoresis of raw and processed flours
In Fig. 1, \(PBF\) raw presented saturation of the main protein groups present in bean, mainly globulins (40–60 kDa) and phytohemagglutinins (27–37 kDa). However, processed \(PBF\) showed weak protein bands, although the same extraction methodology was used. In contrast, \(RRF\) raw and processed showed a similar protein profile, to typical rice grain protein run, such as the presence of albumin (13–110 kDa), α glutenin (30–40 kDa), β glutenin (19–23 kDa) and prolamines (~ 13 kDa). The presence of these proteins indicates their maintenance after processing. \(PBF\) raw showed strong protein bands when compared to \(RRF\) raw and processed results since beans have higher protein content when compared to rice. No legumin band was found, and only a weak vicillin band (~ 47.82 kDa) was observed in the \(PBF\) processed (Fig. 1). These results may be related to the fact that proteins in the 60 kDa range were partially extracted from a flour insoluble complex processed during sample preparation [10]. Phytohemagglutinins or lectins are globular proteins that are present in large quantities in beans. These proteins are highly resistant to proteolytic digestion, the digestive tract is constantly exposed to biologically active lectins contained in fresh and processed [11]. When consumed in large quantities, phytohemagglutinins are toxic. A pre-treatment is necessary to reduce or destroy these proteins [12]. In this study, a weak band in the range of 33.99 kDa was observed in the \(PBF\) processed (Line 3, Fig. 1), indicating that all the processes for obtaining the flour were fundamental for lectin reduction.
3.2 Technological properties
Measurements of cookies' technological properties are shown in Table 1. Under the conditions studied, the levels of \(PBF\) (\({X}_{1}\)) and margarine (\({X}_{2}\)) did not have significant effects (p > 0.05) on the weight reduction (\(WR\)), thickness ratio (\(TR\)), and expansion factor (\(EF\)) of the cookies. However, an increase in \({\text{X}}_{1}\) was able to reduce significantly (p < 0.01) the diameter ratio (\(DR\)) and specific volume (\(SV\)) (Table S1). The effect on these variables can be attributed to the functional properties of \(PBF\) and \(RRF\). Flours with higher protein content increase some properties, like water retention, emulsification, and viscosity of the dough, consequently, occurring an inverse correlation between expansion and volume of dough during baking [8]. The fitted models (Eqs. 1 and 2) had a non-significant \(LoF\) (Table S1, p > 0.05) and the coefficients of determination (\({R}^{2}\)) were acceptable for \(DR\) (0.888) and \(SV\) (0.712), which could be used for predictive purposes.
Table 1
Experimental design and responses for technological properties and sensory analysis of cookies
| Experimental design | | Technological properties | | Sensory analysis |
Trial | \({X}_{1}\) | \({X}_{2}\) | | \(WR\) | \(DR\) | \(TR\) | \(EF\) | \(SV\) | \(H\) | | Taste | Color | Aroma | \(App\) | Texture | \(PI\) |
| g/100 g | g/100 g | | g | cm/cm | cm/cm | cm/cm | mL/g | N | | | | | | | |
1 | 25 | 13.3 | | 1.15 | 1.08 | 1.41 | 5.84 | 1.35 | 1.89 | | 7.30 | 8.09 | 7.33 | 7.96 | 7.47 | 7.17 |
2 | 25 | 19.3 | | 1.18 | 1.12 | 1.32 | 6.38 | 1.32 | 1.12 | | 7.39 | 8.05 | 7.56 | 7.88 | 7.68 | 7.21 |
3 | 75 | 13.3 | | 1.11 | 1.03 | 1.34 | 6.19 | 1.21 | 3.04 | | 6.30 | 7.79 | 6.84 | 7.74 | 6.30 | 5.89 |
4 | 75 | 19.3 | | 1.16 | 1.04 | 1.27 | 6.22 | 1.25 | 1.75 | | 7.00 | 7.74 | 7.25 | 7.58 | 7.28 | 7.02 |
5 | 14.65 | 16.3 | | 1.13 | 1.14 | 1.39 | 6.11 | 1.43 | 1.64 | | 7.32 | 7.53 | 7.47 | 6.98 | 6.96 | 6.79 |
6 | 85.35 | 16.3 | | 1.08 | 1.04 | 1.36 | 5.98 | 1.19 | 1.41 | | 6.74 | 7.42 | 7.12 | 7.32 | 7.04 | 6.02 |
7 | 50 | 12.06 | | 1.06 | 1.06 | 1.45 | 5.62 | 1.31 | 2.72 | | 6.51 | 7.49 | 7.25 | 7.60 | 6.32 | 6.07 |
8 | 50 | 20.54 | | 1.06 | 1.12 | 1.45 | 5.98 | 1.41 | 1.36 | | 7.46 | 7.77 | 7.32 | 7.26 | 7.33 | 7.30 |
9 | 50 | 16.3 | | 1.09 | 1.06 | 1.41 | 5.81 | 1.24 | 1.74 | | 7.00 | 7.88 | 7.37 | 7.86 | 7.33 | 6.58 |
10 | 50 | 16.3 | | 0.97 | 1.06 | 1.35 | 5.99 | 1.28 | 1.70 | | 7.12 | 8.07 | 7.49 | 7.89 | 7.63 | 7.04 |
11 | 50 | 16.3 | | 1.08 | 1.04 | 1.28 | 6.23 | 1.24 | 1.50 | | 7.11 | 7.88 | 7.60 | 7.81 | 7.33 | 7.04 |
12 | 50 | 16.3 | | 1.13 | 1.05 | 1.33 | 5.96 | 1.20 | 1.78 | | 7.25 | 7.63 | 7.39 | 7.60 | 7.39 | 7.39 |
\({X}_{1}\): \(PBF\) addition; \({X}_{2}\): margarine; \(WR\): weight reduction; \(DR\): diameter ratio; \(TR\): thickness ratio; \(EF\): expansion factor; \(SV\): specific volume; \(H\): hardness; \(App\): appearance; \(PI\): purchase intention. Trials of the 9 to 12 are the center points |
$$DR=1.053-0.034{X}_{1}+0.013{{X}_{1}}^{2}+0.017{X}_{2}+0.013{{X}_{2}}^{2}\dots \left(1\right)$$
$$SV=1.258-0.069{X}_{1}+0.043{{X}_{2}}^{2}\dots \left(2\right)$$
The response surface plots for \(DR\) and \(SV\) were constructed using Eqs. (1) and (2) and are shown in Fig. S1 and S2. The regression coefficient of \({X}_{1}\) had a negative impact on \(DR\) (– 0.034) and \(SV\) (– 0.069) values. On the other hand, the \(H\) values ranged from 1.12 to 3.04 N (Table 1). Based on the ANOVA (Table S1), the process variability was explained by the effects, in the following impact order \({X}_{2}\) (59.51%, p < 0.01); \({{X}_{2}}^{2}\) (10.53%, p < 0.05), and \({X}_{1}\) (7.92%, p < 0.05). The model tested (Eq. 3) showed a non-significant \(LoF\) (p > 0.05) and an acceptable \({R}^{2}\) (0.779), which could be used to predict future values within the studied range.
$$H=1.652+0.182{X}_{1}-0.498{X}_{2}+0.229{{X}_{2}}^{2}\dots \left(3\right)$$
The response surface graph for H was constructed using Eq. (3) and is shown in Fig. S3. The addition of \({X}_{1}\) (up to 85.35 g/100 g) caused an increase in \(H\). Some authors observed an increase in hardness when pulses were added to cookies [13, 14]. Incorporating protein-rich flours requires more water to obtain a moldable dough. With this competition and reduction of free water in the dough, it loses consistency and becomes breakable. In addition, the higher the protein content, the higher the water absorption in the dough, consequently, less expansion will make the cookie harder [8, 13]. On the other hand, the addition of \({X}_{2}\) (up to 20.54 g/100 g) reduced \(H\), making the cookies softer (Fig. S3). The addition of fats in the formulation softens the dough and gives it a plastic character. Fat globules cover proteins and starch granules and isolate them. Thus, it prevents the formation of polymers and reduces the density of the net, which reduces the hardness of the cookies [15].
3.3 Sensory analysis
The \(PBF\) addition (\({X}_{1}\)) was able to negatively interfere with the acceptance of the taste, aroma, appearance, and texture, essential criteria for the consumption of cookies (Table 1). Besides, the color, as well as the purchase intention of the cookies, were not influenced by the formulations used in this study. Despite the \(PBF\) content having influenced the acceptance of the cookies, all the evaluated treatments received grades, that on the hedonic scale corresponded from “liked moderately” to “liked very much” (Table 1). On the other hand, considering a relationship between sensory texture and instrumental texture (\(H\)) (– 0.83, p < 0.01, Fig. 2a), it can be noted that there is a rejection by harder cookies, a result achieved by increasing \({X}_{1}\). Also, the addition of margarine (\({X}_{2}\), up to 20.54 g/100 g) had a positive impact on sensory acceptance for taste and texture. Szczygiel et al. [16] reported that 25–35% of \({X}_{1}\) added to the total cracker mass weight had a negative impact on texture and taste.
Trial 1 showed the best acceptance rate (considering all the sensory attributes) while the formulation of trial 3 showed the lowest acceptance rate. However, the rates for almost all analyzed attributes were above 70% for all treatments (Table 1). This reaffirms that, although the increase in \(PBF\) causes less acceptance of the product, the inclusion of a higher content of margarine improves the sensory aspects, making the product more accepted. Baked products require margarine, vegetable shortening, or butter to promote a pleasant taste, incorporate air, increase shelf life, and soften the texture of the final product [17]. Thus, a balance between these ingredients, to serve a product with greater acceptance in the market, must be sought.
3.4 Selection of formulations
The selected formulas were found using multiple desirable characteristics and principal component analysis. For technological properties, optimal process conditions were obtained by overlaying contour plots of \(DR\), \(SV\), and \(H\) over the \({X}_{1}{X}_{2}\) plane (Fig. 2b), to satisfy the technological properties of cookies based on \(PBF\) and \(RRF\) blends. According to the AACC Method 10–54 [18], high-quality cookie flours are usually associated with larger cookie \(DR\), lower \(SV\), and \(H\), thus desirability ranges were: 1.05 ≤ \(DR\) (cm/cm), \(SV\) (mL/g) ≤ 1.32, and 1.5 ≤ \(H\) (N) ≤ 2.0. Figure 2b shows the feasible operating region from the technological standpoint, in the range of 27.5 ≤ \({X}_{1}\) (g/100 g) ≤ 60.0 and 14.2 ≤ \({X}_{2}\) (g/100 g) ≤ 17.5. The cookies obtained in this region guarantee the technological characteristics and necessary attributes in the development of novelty products rich in proteins. On the other hand, the dimensionality reduction of the sensory analysis responses into two principal components (\(PC\)), allowed retaining above 90% of the variation present in the original responses (\(PC\) 1: 64.76% and \(PC\) 2: 26.07%). In the correlation plot (Fig. 2a, c), the responses grouped that presented positive strong correlations (p < 0.01) were: taste-\(PI\) (+ 0.91), texture-\(PI\) (+ 0.83) and taste-texture (+ 0.83). Correlations between taste-aroma, color-\(App\), color-texture, aroma-texture, and aroma-\(PI\) reached intermediate correlations (\(p\) < 0.05). The projection of the trials on the factor map (Fig. 2d) confirms the contribution of all variables in data variability. Trials 1 and 2 are close to the \(CP\) point showing similarity in the results of the technological properties and sensory analysis. It was considered that the increase of \(PBF\) in cookies formulated negatively affected technological and sensory parameters, while the increase in the margarine content improved these aspects. In addition, it can be considered that the excessive increase of fats in the formulations can cause nutritional problems in the long term. Thus, it is important to seek a balance between these ingredients, not forgetting that red rice, also included in the formulation, is a component with good content in fibers and antioxidant elements. The cookies selected to analyze the nutritional composition, bioactive compounds, morphology, and storage were trials 1, 2, and \(CP\). Also, a control treatment with 0 g \(PBF\), 100 g \(RRF\), and 16.3 g margarine/100 g dough was included.
3.5 Chemical composition analysis, iron, and zinc content
Cookies with a higher \(PBF\) addition had higher moisture, ash, protein, iron, zinc, and fiber content as well as lower total carbohydrate content (Table 2). Moreover, trial \(CP\) presented almost 2 times more protein and 3.42 times more fiber than the control formulation. Some authors observed an increase in fiber content with legume incorporation in cookies [19, 20]. Adding pinto beans to cookies not only increases the amount of protein and fiber, but also minerals like iron and zinc (Table 2). Trial \(CP\) had almost 4 times more iron and 1.39 times more zinc content when compared to the control treatment. Rajiv et al. [19] also observed an increase in these minerals when there was a 40 g/100 g substitution of wheat flour for legume flour in comparison to control.
Table 2
Chemical composition (selected and control cookies) and bioactive compounds of pinto beans flour (\(PBF\)) and red rice flour (\(RRF\)) raw and processed, selected cookies and control sample
Component | | Trial (cookies) | | Control | |
1 | 2 | \(CP\) | |
Proximal | | | | | |
Ash (%, db) | 2.43 ± 0.00b | 2.37 ± 0.00b | 2.83 ± 0.05a | 2.05 ± 0.07c | |
Lipid (%, db) | 15.43 ± 0.57b | 22.57 ± 0.61a | 19.48 ± 1.71ab | 16.89 ± 0.23b | |
Protein (%, db) | 9.19 ± 0.24b | 9.89 ± 0.08b | 13.00 ± 0.16a | 5.77 ± 0.08c | |
Carbohydrates (%, db) | 67.14 ± 0.77b | 59.49 ± 0.55c | 56.41 ± 1.68c | 72.90 ± 0.10a | |
Fiber (%, db) | 5.82 ± 0.01b | 5.68 ± 0.01c | 8.28 ± 0.01a | 2.39 ± 0.01d | |
Energetic value (kcal) | 440.44 ± 3.05b | 477.45 ± 3.58a | 445.28 ± 8.02b | 465.24 ± 1.33a | |
Mineral | | | | | |
Iron (mg/100 g) | 1.34 ± 0.02b | 1.40 ± 0.70b | 2.13 ± 0.06a | 0.58 ± 0.13c | |
Zinc (mg/100 g) | 1.42 ± 0.01b | 1.45 ± 0.01b | 1.54 ± 0.03a | 1.11 ± 0.03c | |
Bioactive compounds | | | | |
\(TPC\) (mg \(GAE\)/100 g) | 116.43 ± 1.26b | 122.81 ± 3.75b | 139.46 ± 1.57a | 87.35 ± 0.21c | |
\(AC\) (%) | 71.56 ± 1.08b | 74.96 ± 1.45a | 77.72 ± 1.70a | 70.56 ± 0.26b | |
Tannins (mg \(CE\)/100 g) | 3.47 ± 0.15b | 3.27 ± 0.43b | 4.78 ± 0.10a | 3.28 ± 0.07b | |
| Flours |
Component | \(RRF\) | \(RRF\) | \(PBF\) | \(PBF\) | |
| (Raw) | (Processed) | (Raw) | (Processed) | |
\(TPC\) (mg \(GAE\)/100 g) * | 184.78 ± 5.83a | 79.16 ± 0.68b | 204.65 ± 0.50a | 79.16 ± 0.68b | |
\(AC\)(%) * | 99.92 ± 0.06a | 74.49 ± 1.06b | 95.22 ± 1.57a | 82.16 ± 0.66b | |
Tannins (mg \(CE\)/100 g) * | 17.35 ± 1.24a | 10.00 ± 0.59b | 16.14 ± 0.43a | 7.86 ± 0.01b | |
Values are expressed as the mean ± standard deviation of three replications (n = 3). Cookies made of pinto bean flour (\(PBF\)) and red rice flour (\(RRF\)). Trial 1 (25 g \(PBF\), 75 g \(RRF\), and 13.3 g margarine/100 g dough). Trial 2 (25 g \(PBF\), 75 g \(RRF\), and 19.3 g margarine/100 g dough). Trial \(CP\) (center points’ average) (50g \(PBF\), 50 g \(RRF,\) and 16.3 g de margarine/100 g dough). Trial control (0 g \(PBF\), 100 g \(RRF\), and 16.3 g margarine/100 g dough). db: dry basis; \(TPC\): Total phenolic compound; \(GAE\): Gallic acid equivalent; \(AC\): Antioxidant capacity; \(CE\): Catechin equivalent. Values with different letters in the same row differ statistically in mean using the Tukey test (p < 0.05) and * t-student test (p < 0.05). |
3.6 Bioactive compounds
The process to obtain flours (\(PBF\) and \(RRF\)) retained 43% and 61.4% of phenolic compounds and 74.5% and 86.3% antioxidant capacity, respectively. Only \(PBF\) showed significant retention of tannins (48.7%), losing more than 50% of this component during processing (Table 2). In cookies, it is noted that the increase in the \(PBF\) addition (up to 50 g/100 g) increased the content of total phenolic, tannins, and antioxidant capacity. Some studies also reported an increase in phenolic compounds when beans were added to formulations [21, 22]. However, a reduction in the concentration of total phenols, especially in beans, is expected due to its process of soaking, as these compounds are water-soluble. Additionally, high temperatures used to cook the grains may negatively impact these compounds or even the interaction with proteins, forming complexes [23]. Even with a certain loss after processing, the antioxidant capacity of the flours and cookies was above 70%, which is considered exceptionally good. It is important to highlight that the cooking of raw grains is necessary to eliminate antinutritional substances. Moreover, it increases the digestibility of proteins and starches through the inactivation of proteinase inhibitors, protein denaturation, and starch gelatinization [24]. In addition, the increase in margarine also showed a higher antioxidant capacity (Table 2), this could be due to the addition of antioxidant additives in the margarine.
3.7 Scanning electron microscopy (SEM)
All cookies exhibited homogeneous characteristics, however, with variable sizes of protuberances (Fig. 3). Starch granules are surrounded by a non-gluten-forming protein film and melted and re-solidified fat [25]. The components of the cookie formulations are capable of interaction when ingredients are mixed and kneaded during dough formation, especially in the presence of an emulsifier, such as lecithin. However, it is observed that trial 1 (with less margarine content) presents a different aspect from the other treatments, with less evidence of complexes made with starch, proteins, and lipids (Fig. 3). As expected, the presence of a gluten protein network was not observed in all formulations. However, air bubbles can be observed in various parts of the cookie (Fig. 3). This is due to the formation of gases by chemical agents and the evaporation of water when baked. Higher temperatures increase gas production and water vapor; this expands the dough to go outside, producing air bubbles inside the cookies [9].
3.8 Storage
The water activity (\({a}_{w}\)) increased after day 60 of storage, for all samples. However, these values remained around 0.3. This level of \({a}_{w}\) is important to reduce risks of biological, chemical, and physical deterioration and, consequently, increase shelf-life [26]. The use of margarine may affect reducing \({a}_{w}\) Trial 2 showed a lower \({a}_{w}\), among the other samples (Table 3). On the other hand, the hardness of the cookies was not affected by the storage period (Table 3, p > 0.05), possibly due to the maintenance of low levels of \({a}_{w}\) throughout this period. The cookie samples showed very close pH values, around 6. However, there is a slight increase in pH from day 30, remaining stable until 60 days. At the same time, the samples showed low acidity, especially the control sample. The acidity level did not change significantly (Table 3, p > 0.05). The thermal processing of beans, red rice, and the cookies may have contributed to the stability of these parameters as heat inhibits enzymes that can cause rancidity in whole flours. Regarding the color, there is a darkening of all the cookie samples after day 30, keeping up to 60 days of storage. Cookies tended to become reddish from the same period. The same was observed for the values of hue and saturation. The cookie samples showed more intense colors and within the range of red, after day 30 of storage, remaining stable within up to 60 days. This effect may be due to the additional anthocyanidins released from proanthocyanidins [27]. Also, the formation of anthocyanin-derived pigments stabilizes the flavylium red-colored form [28]. At the end of the storage, all samples, including the control sample, showed similar values for chroma a* and b*, hue, and saturation (Table 3). Changes in product color during storage indicate changes in freshness shelf life and therefore play an important role in food selection and purchase [29].
Table 3
Storage analysis of selected cookies and control sample
Component | Day | | Trial (cookies) | | Control |
1 | 2 | \(CP\) |
\({a}_{w}\) | 1 | 0.15 ± 0.01cB | 0.15 ± 0.02bB | 0.22 ± 0.01cA | 0.17 ± 0.01cAB |
30 | 0.22 ± 0.01bAB | 0.19 ± 0.01bB | 0.24 ± 0.01bA | 0.23 ± 0.01bAB |
60 | 0.32 ± 0.01aB | 0.32 ± 0.01aB | 0.35 ± 0.01aAB | 0.36 ± 0.01aA |
\(H\) | 1 | 1.95 ± 0.49aA | 1.42 ± 0.47aA | 2.71 ± 1.65aA | 1.57 ± 0.87aA |
30 | 2.31 ± 0.82aAB | 1.40 ± 0.58aB | 2.80 ± 1.49aA | 1.65 ± 0.55aAB |
60 | 2.53 ± 0.75aA | 1.35 ± 0.45aAB | 1.96 ± 1.03aAB | 1.46 ± 0.63aB |
pH | 1 | 6.52 ± 0.03bA | 6.28 ± 0.06bB | 6.44 ± 0.06bAB | 6.50 ± 0.06bAB |
30 | 6.83 ± 0.01aA | 6.71 ± 0.01aC | 6.73 ± 0.01aC | 6.79 ± 0.01aB |
60 | 6.71 ± 0.04aA | 6.66 ± 0.04aA | 6.72 ± 0.03aA | 6.77 ± 0.09abA |
\(TTA\) | 1 | 0.18 ± 0.01aB | 0.17 ± 0.01bC | 0.21 ± 0.01aA | 0.14 ± 0.01aD |
30 | 0.19 ± 0.01aAB | 0.21 ± 0.01aA | 0.22 ± 0.01aA | 0.17 ± 0.02aB |
60 | 0.16 ± 0.01bB | 0.19 ± 0.01abA | 0.21 ± 0.01aA | 0.15 ± 0.01aB |
L* | 1 | 39.26 ± 0.62bA | 39.64 ± 0.52bA | 38.25 ± 1.74bA | 38.36 ± 0.84bA |
30 | 44.76 ± 0.52aA | 44.63 ± 0.49aA | 44.11 ± 0.09aA | 44.75 ± 0.18aA |
60 | 44.34 ± 0.73aA | 43.59 ± 1.06aA | 44.38 ± 0.41aA | 43.80 ± 0.25aA |
a* | 1 | 7.31 ± 0.20bA | 6.58 ± 0.10bA | 7.36 ± 0.74aA | 6.70 ± 0.24bA |
30 | 8.14 ± 0.07aA | 7.76 ± 0.02aA | 7.90 ± 0.23aA | 7.26 ± 0.18abB |
60 | 8.25 ± 0.29aA | 7.87 ± 0.69aA | 8.13 ± 0.16aA | 7.61 ± 0.54aA |
b* | 1 | 4.57 ± 0.20bA | 3.85 ± 0.16bB | 3.55 ± 0.30bB | 3.55 ± 0.30bB |
30 | 8.19 ± 0.14aA | 8.10 ± 0.16aA | 7.81 ± 0.37aA | 6.85 ± 0.35aB |
60 | 8.38 ± 0.31aA | 8.12 ± 1.38aA | 8.41 ± 0.27aA | 7.36 ± 0.84aA |
Hue | 1 | 0.56 ± 0.02bA | 0.53 ± 0.02bAB | 0.45 ± 0.06bB | 0.49 ± 0.03bAB |
30 | 0.79 ± 0.01aA | 0.81 ± 0.01aA | 0.78 ± 0.01aAB | 0.76 ± 0.01aB |
60 | 0.79 ± 0.00aA | 0.80 ± 0.04aA | 0.80 ± 0.01aA | 0.77 ± 0.02aA |
C* | 1 | 9.59 ± 0.27bA | 8.51 ± 0.03bB | 9.14 ± 0.69bAB | 8.48 ± 0.36bB |
30 | 12.23 ± .010aA | 11.81 ± 0.06aA | 11.81 ± 0.41aA | 10.68 ± 0.35aB |
| 60 | 12.44 ± 0.44aA | 11.93 ± 1.38aA | 12.33 ± 0.26aA | 11.29 ± 0.96aA |
Values are expressed as the mean ± standard deviation of three replications (n = 3). \({a}_{w}\): water activity; \(H\): hardness; \(TTA\): titratable total acidity; L*: luminosity; a*: variation from green to red; b*: variation from blue to yellow; C*: saturation. Cookies made of pinto bean flour (\(PBF\)) and red rice flour (\(RRF\)). Trial 1 (25 g \(PBF\), 75 g \(RRF\), and 13.3 g margarine/100 g dough). Trial 2 (25 g \(PBF\), 75 g \(RRF\), and 19.3 g margarine/100 g dough). Trial \(CP\) (center points’ average) (50g \(PBF\), 50 g \(RRF\), and 16.3 g de margarine/100 g dough). Trial control (0 g \(PBF\), 100 g \(RRF\), and 16.3 g margarine/100 g dough). Values with different lowercase letters in the same column differ statistically in mean using the Tukey test (p < 0.05). Values with different uppercase letters in the same row differ statistically in mean using the Tukey test (p < 0.05) |