Development of functional low glycemic index flakes using white sweet potatoes and precooked peas

Abstract

Variations in precooked peas (Pisum sativum L.) and white sweet potatoes (Ipomoea batatas (L.) Lam), which exhibited different characteristics in nutritional composition, might affect the functional properties of food. This study aimed to determine the optimal proportion of white sweet potato flour and precooked pea flour in composite flour based on physicochemical characteristics and sensory acceptance. The flakes were then evaluated for glycemic index (GI) and glycemic load (GL). Flakes were prepared using a fixed proportion of wheat flour (35%) combined with varying ratios of white sweet potato flour to precooked pea flour: F1 (30%:35%), F2 (35%:30%), and F3 (40%:25%). Peas were precooked by autoclaving prior to drying and milling into flour, while white sweet potatoes were directly processed into flour. The resulting flours were blended into composite formulations, which were then processed into dough and subsequently steamed and baked to produce flakes. Analyses included proximate composition, dietary fiber, amylose, untargeted metabolomic profiling, and hardness. Sensory evaluation was conducted using a hedonic test. Increasing precooked pea flour significantly increased ash, fat, protein, hardness, amylose, dietary fiber, and sensory attributes (p < 0.05). F3 showed the lowest fat content, highest dietary fiber (>4.8%), and highest acceptability (4.1/5). Untargeted metabolomic analysis identified fifteen phytochemicals (VIP > 1.0). Overall, precooked pea substitution enhanced protein, fiber, resistant starch, and bioactive compounds, while higher white sweet potato levels improved carbohydrate content, texture, sensory acceptance, and umami-related compounds. Both F1 and F3 exhibited low GI values (32 and 49), indicating potential as functional flakes for glycemic control.

Keywords:
Flakes; Metabolomic; Nutrition; Pea; Precooked; Sensory; White sweet potato

HIGHLIGHTS

Increasing precooked pea flour enhanced protein, dietary fiber, resistant starch, and hardness of flakes

Higher white sweet potato proportion improved texture, sensory acceptance, and umami-related metabolites

Composite flour ratios significantly influenced physicochemical and metabolomic profiles of flakes

The optimal formulation (35:40:25) achieved high fiber (>4%) and resistant starch (>6%) with good consumer acceptability

1 Introduction

The increasing consumption of carbohydrates is one of the contributing factors to the rise of non-communicable diseases (NCDs), particularly diabetes mellitus (DM). Therefore, food innovation, especially the development of healthier flake products, is needed as part of dietary management strategies to prevent NCDs (Low et al., 2015). The advancement of industrial innovation, accompanied by changes in lifestyle, has further emphasized the demand for improved food quality. Flake products have gained substantial attention as convenient, shelf-stable, and widely acceptable foods across age groups. However, commercial flakes are predominantly produced from refined grains, resulting in high glycemic responses and limited micronutrient diversity. The development of flakes from composite flour of white sweet potatoes and pre-cooked peas offers an opportunity to enhance nutritional quality, improve dietary diversity, and provide a functional food option with potential metabolic benefits.

White sweet potatoes are an important carbohydrate source, with starch as the major component. The starch of white sweet potatoes contains approximately 35.99% amylose (Saman et al., 2020). Starch with high amylose content (>25%) is generally digested more slowly due to its linear and less branched molecular structure, which limits enzymatic hydrolysis and results in a lower glycemic response (Behall & Hallfrisch, 2002). Consistent with this property, white sweet potatoes have been reported to exhibit a glycaemic index (GI) of 54 (Dwi & Rohmawati, 2019), which is classified as a low-GI food (<55) (Peres et al., 2023). Furthermore, legumes such as peas also demonstrate favorable glycemic responses. For example, foods formulated with peas have been reported to reduce postprandial blood glucose levels in individuals with type 2 diabetes from 10 mmol/L to 6.8 mmol/L within 30 minutes (Winham et al., 2025). Therefore, combining white sweet potatoes and peas in a composite formulation may provide a promising low-GI food alternative that could support the prevention or management of DM. Previous studies have consistently reported that dietary patterns characterized by low glycemic index and low glycemic load are associated with improved metabolic outcomes, particularly among individuals with impaired glucose metabolism (Howlett & Ashwell, 2008).

Beyond their carbohydrate characteristics, white sweet potatoes also contain several bioactive compounds that may contribute to metabolic health. White sweet potatoes have antioxidant activity around 2.72 μmol TE/g (Teow et al., 2007). Furthermore, metabolomic profiling has shown that sweet potato varieties, including white sweet potatoes, contain several essential amino acids, including L-valine, L-tryptophan, L-isoleucine, L-phenylalanine, methionine, and L-lysine (Wan et al., 2024). These amino acids have been associated with antioxidant activity and may help counteract oxidative stress commonly observed in DM (Martínez et al., 2017). In addition to amino acids, sweet potato varieties contain diverse organic acids, including citric acid, pyruvic acid, succinic acid, and quinic-acid derivatives. In a comparative metabolomic analysis, citric acid levels in white sweet potato flesh were reported to be approximately 6.24-fold higher than in orange sweet potato, while several quinic-acid derivatives were also identified as differential metabolites among sweet potato varieties (white sweet potatoes, orange sweet potatoes and purple sweet potatoes) (Wan et al., 2024).

Peas are widely recognized as a valuable source of plant protein and bioactive compounds. However, the incorporation of pea ingredients into food products may introduce certain sensory challenges, particularly bitter taste and characteristic beany flavor. Correia et al. (2025) reported that peas contain at least 22 amino acids, peptides, and related compounds associated with bitter sensory attributes, along with several phenolic compounds that may contribute to antioxidant activity. These sensory characteristics are often associated with protein composition and lipid oxidation processes in legumes. Processing methods applied during product development can significantly influence these attributes. Thermal treatment, such as autoclaving used in producing pre-cooked peas, can alter protein structures and reduce conformational flexibility. A systematic review by Lao et al. (2024) reported that thermal processing decreases β-structure content and increases protease accessibility, which can enhance protein digestibility by breaking proteins into smaller peptides and amino acids. These peptides may influence flavor development, as some protein hydrolysis products contribute to taste attributes such as bitterness or umami, thereby affecting overall sensory acceptance.

Thermal processing may influence lipoxygenase (LOX) activity, which is another factor contributing to the beany flavor of legumes (Lao et al., 2024). Therefore, in composite flakes formulated from different proportions of white sweet potatoes and precooked peas, the interaction between pea-derived amino acids, sweet potato carbohydrates, and thermal processing conditions plays an important role in determining both flavor development and nutrient content of product.

Processing conditions on precooked peas flour that involve autoclaving and cooling have been shown to increase dietary fiber content by up to 42.25% and resistant starch content by 35.1% compared with conventional boiling (Mustikaningrum et al., 2013). In addition, a recent review reported that white sweet potato flour contains approximately 20% to 30% dietary fiber (Islam, 2024). Different levels of fiber in precooked peas and white sweet potatoes might affect the dietary fiber of flakes. Increased dietary fiber levels may influence the structural properties of cereal products by reducing expansion and increasing product density, which can lead to harder textures (Brennan et al., 2008). Protein and moisture levels also play important roles in determining the texture of baked products. During baking, moisture evaporation creates air pockets within the product matrix, contributing to the development of crispness (Ayuningtyas & Aan, 2025). Therefore, variations in the ratio of white sweet potato and pea flour in composite formulations may influence fiber content, protein levels, and moisture distribution, which in turn affect the hardness, crispiness, and overall sensory acceptability of the flakes.

Despite the nutritional potential of white sweet potatoes and pre-cooked peas, scientific evidence regarding composite flakes derived from these ingredients remains limited. Previous studies have not systematically examined optimal formulation ratios, the influence of processing conditions on starch digestibility and nutrient retention, or the resulting glycemic response of such products. In addition, information on the metabolomic characteristics and sensory acceptability of flakes formulated from these ingredients is still scarce. Addressing these gaps is essential for understanding how ingredient composition and processing interact to determine the nutritional, metabolic, and sensory properties of flake products. Therefore, this study aims to evaluate composite flakes formulated from white sweet potatoes and pre-cooked peas by analyzing their formulation characteristics, nutritional composition, metabolomic profiles, glycemic response, and sensory properties. To our knowledge, this is the first study to comprehensively investigate the combined effects of composite formulation of white sweet potato and pre-cooked pea flours on the metabolomic profile, glycemic index response, and sensory quality of flake products.

2 Materials and methods

2.1 Materials

Peas and white sweet potatoes used in this study were obtained from UD, Tani Maju, a farm grocery in Yogyakarta.

2.2 Precooked peas and white sweet potato flour preparation

Fresh white sweet potatoes were peeled and cleaned, then washed under running water and cut into small pieces (approximately 2-3 cm thickness) to facilitate uniform drying. The sliced samples were then dried in a hot-air oven at 70 °C for 4 hours. The dried sheeted sweet potatoes were cooled to room temperature and subsequently ground using a grinder (Stainless Steel Disintegrator, China). The resulting powder was sieved with a sieve shaker (Galfindo Sieve, Indonesia) through a 60-mesh to ensure uniform particle size. The flour obtained was packed in airtight containers and stored at refrigerator (4 ± 1 °C) until further analysis.

Peas seeds were soaked in a sodium bicarbonate food grade (Kopoe-kopoe, Indonesia) solution with a water-to-pea ratio of 3:1 (v/w) to facilitate hydration and reduce anti-nutritional factors in room temperature 26 °C to 27 °C, with 65% RH for 24 hours. The soaked peas were then subjected to thermal processing using an autoclave (B-one Analog AA, Indonesia) at 121 °C and 15 psi for 145 s. Following autoclaving, the samples were dried in a hot-air oven (Labtech, Italy) at 50 °C for 20 hours (moisture content <9%). The pea seeds were subsequently ground using a grinder to obtain a fine powder, then sieved in a 60-mesh sieve shaker. The resulting flour was collected as pre-cooked pea flour and stored in airtight containers in a refrigerator (4 ± 1 °C) until further use.

2.3 Composite flour proportion

The proportion of wheat flour: white sweet potatoes: pre-cooked pea flour that was used for making the flakes was based on a preliminary study that shows the proportion of wheat flour: white sweet potatoes: pre-cooked pea flour (25: 45: 30) has better acceptance and water absorbance compared to 25: 55:20. Therefore, the main formulations (wheat flour: white sweet potatoes: pre-cooked peas flour) that were used in the study were (1) 35:30:35, (2) 35:35:30, and (3) 35:40:25. Wheat flour used in this study was commercial flour (Segitiga Biru, Bogasari Flour Mills, Indonesia). Each formulation will be tested for proximate, amylose, dietary fiber, resistant starch, amylose, hardness, and organoleptic.

2.4 Flakes preparation

For the preparation of flakes, all composite formulations added 2 g of salt, 10 g of commercial diabetic friendly sweetener (Diabetasol, Indonesia), and 65 mL of water. The table of flakes recipe (in 100 g) can be seen in Table 1.

Table 1
Flakes recipe.

Formulation	Wheat Flour: white sweet potato flour: pre-cooked pea flour (%)	Water (%)	Salt (%)	Commercial diabetic friendly sweetener (%)
F1	35 : 30 : 35	65	2	10
F2	35 : 35 : 30	65	2	10
F3	35 : 40 : 25	65	2	10

The dough of all formulations was sheeted using a pasta maker (Shuma Bello 150AT, Indonesia) and adjusted to a uniform thickness of 3 mm, length 2 cm, and width 2.5 cm. The sheeted dough is then steamed at 100 °C for 10 minutes with a lid to cover the pan. After that, put the sheeted dough at room temperature (27 °C). The sheeted dough was then flattened with a roller and shaped to a thickness of 0.5 mm, then baked in an oven (Idealife, Indonesia) at 150 °C for 10 minutes to be flakes.

2.5 Proximate analysis

Proximate analysis was moisture, ash, fat, carbohydrate, and protein. Proximate analysis conducted according to the Association of Official Analytical Chemists (1995).

2.5.1 Moisture

Moisture content was determined using the oven-drying method according to AOAC method 925.10 (Association of Official Analytical Chemists, 1995). Stainless steel oven dishes were cleaned, then dried in an oven at 100 °C for 1 hour to obtain a constant weight. After cooling in a desiccator, the dishes were weighed. Two grams of each sample were placed in the dishes and dried at 100 °C until constant weight was reached. The dishes and contents were cooled in a desiccator and reweighed (Equation 1).

$MoistureContent\left(\%\right)=\frac{\left(W2-W3\right)}{\left(W2-W1\right)}x100$ (1)

where: W1 = W1 = Weight of empty dish, W2 = Weight of dish + sample before drying and W3 = Weight of dish + sample after drying.

2.5.2 Ash content

Ash content was determined according to AOAC method 923.03 (Association of Official Analytical Chemists, 1995). Two grams of the sample were placed in a silica dish that had been previously ignited, cooled, and weighed. The dish was first gently ignited and then heated in a muffle furnace (Suhatherm, Indonesia) at 550 °C for 3 hours until white or grey ash was formed. After cooling in a desiccator (Duran Desiccator Vacuum Glass), it was weighed (Equation 2).

$AshContent\left(\%\right)=\frac{\left(W2-W3\right)}{\left(W2-W1\right)}x100$ (2)

where: W1 = W1 = Weight of empty dish, W2 = Weight of dish + sample before ashing and W3 = Weight of dish + sample after ashing.

2.5.3 Fat content

Fat content was determined using the Soxhlet extraction method according to AOAC method 920.39 (Association of Official Analytical Chemists, 1995). Fat was determined using the Soxhlet extraction (micro Soxhlet Pyrex, Germany). A 500 mL round-bottom flask was filled with 300 mL of petroleum ether (Smart-Lab, Indonesia) and connected to the Soxhlet extractor. Two grams of the sample were placed into a labelled thimble, which was sealed with cotton wool. The system was refluxed for six hours. After extraction, the thimble was removed, and the ether was recovered. The flask was dried in an oven (Memmert, Germany) at 105 °C for 1 hour, cooled in a desiccator, and weighed (Equation 3).

$FatContent\left(\%\right)=\frac{Weightoffat}{Weightofsample}x100$ (3)

2.5.4 Carbohydrate content

Total carbohydrate content was determined by difference according to AOAC guidelines, calculated by subtracting the sum of moisture, protein, fat, and ash from 100%.

2.5.5 Protein content

Protein content was determined using the Kjeldahl Method according to AOAC method 979.09 (Association of Official Analytical Chemists, 1995). Approximately 0.5-1.0 g of sample was digested with concentrated H₂SO₄ (Smart-Lab, Indonesia) and a catalyst at ~420 °C until clear, followed by distillation with NaOH (Smart-Lab, Indonesia) (40-50%) using a semi-automatic Kjeldahl unit (Velp Scientifica, Italy). The released NH₃ (Smart-Lab, Indonesia) was trapped in 4% boric acid and titrated with 0.1 N HCl (Smart-Lab, Indonesia). A blank was included, and protein content was calculated using a conversion factor of 6.25.

2.6 Dietary fiber, resistant starch and amylose analysis

2.6.1 Dietary fiber analysis

Dietary fiber content was analyzed using the enzymatic-gravimetric method based on AOAC Official Method 991.43 (Association of Official Analytical Chemists, 1995). Approximately 0.5 ± 0.005 g of sample was mixed with 40 mL MES-TRIS buffer (Sigma-Aldrich, USA) (pH 8.2), treated with α-amylase (Sigma-Aldrich, USA), and incubated at 95 °C to 100 °C for 35 minutes. After cooling to 60 °C and rinsing the beaker walls with distilled water, protease was added and incubated at 60 °C for 30 minutes. The mixture was then acidified with 0.561 N HCl (Smart-Lab, Indonesia) to pH 4.1-4.6, treated with amyloglucosidase (Sigma-Aldrich, USA), and incubated again at 60 °C for 30 minutes. The digest was precipitated with 225 mL of 95% ethanol (Sigma-Aldrich, USA) and allowed to stand for 1 hour, then filtered through a pre-weighed ashless filter paper (No. 42). The residue was washed with 78% ethanol (Sigma-Aldrich), 95% methanol (Sigma-Aldrich, USA), and acetone (Sigma-Aldrich, USA) (each twice), then dried with oven (Memmert Laboratory, Germany) to constant weight under vacuum at 105 °C. Protein and ash corrections were determined from both sample and blank residues, and total dietary fiber (%TDF) was calculated using corrected residue weight relative to sample mass.

2.6.2 Amylose analysis

Amylose analysis was conducted according to Chen et al. (2025). The sample was dispersed in sodium hydroxide (Smart-Lab, Indonesia) solution to ensure complete starch gelatinization. The mixture was subsequently neutralized prior to the addition of an iodine reagent, which selectively binds to linear amylose chains and forms a blue-colored complex. The absorbance of the resulting solution was measured using a spectrophotometer using an Ultraviolet-Visible (UV-VIS) spectrophotometer (UV160, Shimadzu, Japan) at 620 nm.

2.6.3 The resistant starch analysis

The resistant starch analysis used enzymatic methods with the Resistant Starch Rapid Assay Kit Megazyme. A 100 mg sample was placed into a centrifuge tube (Eppendorf 5810R, USA) and mixed with 3.5 mL of sodium maleate buffer (pH 6.0). The tube was then placed in a water bath (GFL 1002, Germany) at 37 °C for 5 minutes, followed by the addition of 0.5 mL of pancreatic α-amylase and amyloglucosidase solution. The mixture was incubated for 4 hours in a shaking water bath (Labtech LSB 045S, Indonesia). Subsequently, 4 mL of 95% ethanol (Smart-Lab, Indonesia) was added, and the tube was centrifuged at 4000 rpm for 10 minutes. The supernatant was discarded, and the pellet was washed twice with 2 mL of 50% ethanol (Smart-Lab, Indonesia). The resulting pellet was used for resistant starch analysis. The pellet was mixed with 2 mL of 1.7 M sodium acetate (Labotiq, Indonesia), and the suspension was placed in a shaking water bath (Labtech, Italy), at 4 °C for 20 minutes. Next, 8 mL of 1 M sodium acetate buffer (pH 3.8) and 1 mL of AMG (3300 U/mL) were added. The suspension was incubated in a water bath at 50 °C for 3 minutes. An aliquot of 0.1 mL of the supernatant was reacted with 3 mL of glucose oxidase/peroxidase (GOPOD) reagent in a test tube and incubated at 50 °C for 20 minutes. The mixture was then transferred into a 100 mL volumetric flask and diluted to volume with distilled water. Absorbance was measured at 510 nm using a UV-VIS spectrophotometer (UV160, Shimadzu, Japan).

2.7 Reducing sugar analysis

Reducing sugar content was determined following a modified Nelson-Somogyi method. A total of 1 g (or 1 mL) of homogenized sample was weighed and transferred into a 100-mL beaker, followed by the addition of 30 mL of distilled water. Several drops of lead acetate were added until the solution became clear. The mixture was transferred into a 100 mL volumetric flask and diluted to the mark with distilled water, then filtered through filter paper. Sodium oxalate (Labbox) was added in an amount equivalent to the previously added lead acetate to remove excess lead ions. The resulting filtrate was stored for further dilution when necessary. An aliquot of 1 mL sample solution was pipetted into a test tube, followed by the addition of 1 mL Nelson reagent and 1 mL arsenomolybdate reagent, which was prepared from ammonium molybdate tetrahydrate (Merck, Germany), sodium arsenate dibasic heptahydrate (Smart-Lab, Indonesia), sulfuric acid (Merck, Germany), and 7 mL distilled water. The absorbance of the developed color was measured (Shimadzu UV-1800 with 540 nm, Japan), and reducing sugar (%) was calculated using Formula 4:

$ReducingSugar\left(\%\right)=\frac{\left(XxDFxV\right)}{SampleWeight\left(mg\right)}x100$ (4)

where: X is the concentration obtained from the standard curve, DF is the dilution factor, and V is the sample volume used.

2.8 Hardness analysis

The hardness analysis was measured in the final baked product using a texture analyzer (LLOYD Instruments, TA plus AMETEK, UK), calibrated to a testing velocity of 0.5 mm/s, compressing the sample to 50% of its height and a wait time of 0.5.

2.9 Sensory analysis

Sensory evaluation was conducted using a 5-point hedonic scale, where 1 = dislike it very much, 2 = dislike it, 3 = neither like nor dislike it, 4 = like it, and 5 = like it very much. The 30 untrained panelists were used for sensory analysis with clearance ethics number 1.583/VIII/HREC/2023. The age of the panelists was between 20-23 years old.

2.10 Metabolomic profile analysis

The instrument used in this analysis was: Thermo Scientific™ Dionex™ Ultimate 3000 RSLCnano Ultra-High-Performance Liquid Chromatography (UHPLC) coupled with Thermo Scientific™ Q Exactive™ High Resolution Mass Spectrometer with mobile Phase: A Water + 0.1% Formic Acid, mobile phase B Acetonitrile + 0.1% Formic Acid. The analytical Column: Phenyl Hexyl 100 mm x 2, with flow: 0.20 mL/min, and the injection volume sample was 5 µL. Run time: 30 min. The resolution used was 70,000 FWHM, data-dependent MS2 at 17,500 FWHM, heated Electrospray Ionization (H-ESI) positive, negative Compound Identification by Thermo Scientific™ Compound Discoverer Software. Metabolite extraction was carried out using a solid sample. The solvent mixture consisted of methanol, acetonitrile, and water in a ratio of 4:4:2. The sample was finely ground using a mortar, and approximately 25 mg of the homogenized sample was weighed and mixed with 1 mL of the solvent mixture. The suspension was vortexed for 3 minutes, followed by sonication for 60 minutes to enhance metabolite extraction. After sonication, the sample was vortexed again for 3 minutes to ensure complete mixing. The mixture was then centrifuged at 5000 rpm for 5 minutes, and the resulting supernatant was filtered through a 0.2 µm Millex filter. Finally, 5 µL of the filtrate was injected into the High-Resolution Mass Spectrometry (HRMS) (Thermo Fisher Scientific, Bremen, Germany) system for analysis. Compounds were identified based on best-match discoveries using the mzCloud database. The processed data were subsequently analyzed using MetaboAnalyst-6 software. Before analysis, data were normalized by sum, log-transformed (base 10), and filtered using the relative standard deviation (RSD = SD/mean) criterion. The resulting dataset was then scaled using the auto-scaling technique. Partial Least Squares Discriminant Analysis (PLS-DA) was performed, and the results were interpreted based on Variable Importance in Projection (VIP) values.

2.11 Glycemic Index (GI) and Glycemic Load (GL) analysis

The GI and GL analyses were conducted under ethics no 5672/B.1/KEPK-FKUMS/III/2025. The 12 subjects that used in this study must predetermined inclusion criteria such as having a normal Body Mass Index (BMI) of 18.5-22.9 kg/m²; aged between 21-39 years, not diagnosed with DM or impaired glucose tolerance, non-smokers, not under medication, not following any special diet, and with no history of chronic disease and required to fasting for 10-12 hours before testing (water permitted). In the control test, each panelist was given white bread containing 50 g of available carbohydrates (day 1). Blood samples were collected at 0, 15, 30, 45, 60, 90, and 120 minutes after consumption, and after 2 days, the subject should fast for 10-12 hours before testing (water permitted). The flakes that were given to the panelist were 37.64 g, equivalent to 50 g of available carbohydrates/serving. Blood sugar was tested at 0, 15, 30, 45, 60, 90, and 120 minutes. Blood glucose levels were measured using a glucometer (Easy Touch, Taiwan), and the results were recorded at each sampling time. The amount of test food was determined based on Suryaningrum & Rustanti (2016) and its available carbohydrate content using the following Equations 5 and 6:

$\text{Available Carbohydrate}=\left(1.1\text{x starch}\right)+\text{carbohydrates by difference}$ (5)

$\text{Amount of test food}\left(\text{g}\right)=\frac{50\text{gr available carbohydrate}}{\text{available carbohydrate content of the sample}}\text{x}100$ (6)

The total starch of F1 flakes was 45.57%, and F3 50.07%. The amount of test food F1 flakes was 37.64 g, while F3 flakes were 35.83 g. The GI was calculated using the Incremental Area Under the Curve (IAUC) method, as shown below (Equations 7 and 8):

$\text{GI}=\frac{\text{iAUC of test food}}{\text{iAUC of reference food}}x100$ (7)

$\text{GL=}\frac{\text{GI x carbohydrate content per serving}\left(\text{g}\right)}{\text{100}}$ (8)

2.12 The statistical analysis (sentence case)

The correlation between different composite formulations on sensory acceptance was analyzed with Kruskal Wallis followed by Mann-Whitney. The correlation between different composite formulations on moisture, protein, fat, dietary fiber, amylose, and hardness was analyzed using SPSS 21 with Analysis of Variance (ANOVA) test followed by Least Significant Difference (LSD) test with significance P< 0.05, while the relationship between different compositions of formulation with total carbohydrate was analyzed with Kruskal-Wallis followed by Mann-Whitney test.

3 Results and discussion

3.1 Proximate analysis

Table 2 shows that different formulations of white sweet potatoes and precooked peas affect the water, protein, ash, and lipid content of flakes. Increasing the level of white sweet potatoes significantly reduced the water content, protein, fat, and ash content. The formula F3 had the lowest water, protein, fat, and ash content compared to formula F1 and F2.

Table 2
Proximate analysis of flakes.

Formulations	Moisture Content (%) (mean ± SD)	Protein Content (%) (mean ± SD)	Fat Content (%) (mean ± SD)	Ash Content (%) (mean ± SD)	Total Carbohydrate (%) (mean ± SD)
F1	6.44 ± 0.93a	7.23 ± 0.03c	0.83 ± 0.09b	3.90 ± 0.03c	82.72 ± 0.61a
F2	5.35 ± 0.60b	6.76 ± 0.09b	0.47 ± 0.11a	3.50 ± 0.07a	83.33 ± 1.03a
F3	4.77 ± 0.16b	6.27 ± 0.11a	0.39 ± 0.13a	3.61 ± 0.10b	84.46 ± 0.14b
p-value	0.02	<0.01	<0.01	<0.01	0.04

• The formulation of composite flour (wheat: sweet white potatoes, pre-cooked peas) was F1 (35:30:35), F2 (35:35:30) and F3 (35:40:25). Moisture, protein, lipid, and ash content were analysed with One_way ANOVA followed by the LSD test, while total carbohydrate was analysed with Kruskall Wallis followed by the Mann-Whitney test. Different superscript letters indicate significant differences at a significance level of p < 0.05.

Based on Indonesia Standard Food (SNI)1-2886-2000, it could be mentioned that the maximum level of moisture and ash content was 4%, while the minimum protein content was 5%, fat content was a maximum of 30%, and the carbohydrate was a minimum of 60% (National Standardization Agency, 2000). From the data, it can be seen that the ash, protein, fat, and carbohydrate contents already met the requirements of flakes based on the Indonesian Food Standard. However, the moisture content of all formulations of flakes was more than 4%. Thus, all flakes might have a short expiration date. The best flake based on moisture content was F3, since it is closer to Indonesian Standard food flakes. Furthermore, the moisture content of all flakes was less than 8%. According to Akther et al. (2020), the authors mentioned that a dry product that contained less than 8% moisture content was considered ideal because it can prevent the growth of microorganisms in food. This study shows that increasing precooked peas substitution enhanced the moisture content, fat, protein, and ash. Increasing moisture content was in line with protein content in F1 flakes that contained a higher number of precooked peas. Heat treatment causes protein molecules to denature and form a compact network of fibers or matrices, trapping water within the protein molecules and preventing their release (Totosaus et al., 2002). Consequently, as the proportion of pre-cooked pea flour increases, the amount of bound water also increases, leading to a higher moisture content in the flakes.

This study also found that increasing the level of precooked pea substitution significantly increased the protein content of the flakes. This is likely due to the higher protein composition of precooked peas compared with white sweet potatoes. Mustikaningrum et al. (2013) reported that precooked peas contain 27.3% protein, whereas a study by Paula et al. (2021) showed that white sweet potatoes contain only 3.19 to 4.10% protein. All flake formulations contained more than 5% protein, indicating that they met the minimum requirement for a food product to be classified as a protein source, which is at least 5 g/100 g (Food and Agriculture Organization, 1997). In contrast, the carbohydrate content of the flakes decreased significantly as the proportion of white sweet potato was reduced. This is consistent with the fact that precooked peas contain lower carbohydrate levels than white sweet potatoes, as reported by Mustikaningrum et al. (2013), who found carbohydrate contents of 54.08% in precooked peas and 76.42% in white sweet potatoes.

3.2 Total dietary fiber, resistant starch, and amylose content

The average dietary fiber, resistant starch, and amylose content of flakes were given in Table 3. There was a significant correlation between different composite flours and total fiber, with a significance P-value <0.01. Formulations F1 and F3 had a higher total fiber content compared to F2. Based on the data in Table 3, it can be seen that there was no significant difference in resistant starch across all formulations, while the ratio of white sweet potato and precooked peas significantly affects amylose content in flakes, with a P-value of 0.01. Formula F2 had an amylose content of 19.56%, while Formulas 1 and 3 each had an amylose content of 18.37%. Formula F2 had the highest content of amylose, that could be that increasing the number of precooked peas increases the amount of amylose.

Table 3
Total dietary fiber, resistant starch and amylose content of flakes.

Formulations	Soluble Fiber content (%) (mean ± SD)	Insoluble Fiber content (%) (mean ± SD)	Total Fiber content (%) (mean ± SD)	Resistant Starch (%) (mean ± SD)	Amylose content (%) (mean ± SD)
F1	0.44 ± 0.06a	4.37 ± 0.05a	4.81 ± 0.10a	6.02 ± 0.18	18.65 ± 0.34b
F2	0.26 ± 0.04b	4.35 ± 0.09a	4.61 ± 0.09b	6.70 ± 0.30	19.56 ± 0.14a
F3	0.36 ± 0.03c	4.55 ± 0.06b	4.91 ± 0.07a	6.21 ± 0.58	18.37 ± 0.68b
p-value	<0.01	<0.01	<0.01	0.116	0.01

• The formulation of composite flour (wheat: sweet white potatoes, pre-cooked peas) was F1 (35:30:35), F2 (35:35:30) and F3 (35:40:25). Different superscript letters indicate significant differences at a significance level of p < 0.05.

The study shows that increasing the number of white sweet potatoes to 40% increases the amount of insoluble fiber. The highest level of insoluble dietary fiber was observed in formulation F3 (4.55%), while the lowest was again in F1 (4.35%). This might suggest that white sweet potatoes might contain more insoluble fiber compared to precooked peas. This theory is supported by a study that mentioned that sweet potatoes contained greater concentrations of hemicellulose (4.94%), cellulose (12.84%), and lignin (17.41%) than pigeon peas (Kaur et al., 2024; Vannini et al., 2021). Interestingly, although F3 contained a higher proportion of white sweet potatoes (40%), its total dietary fiber content was not significantly different from F1, indicating a non-linear relationship between formulation and measured fiber content. This suggests that matrix interactions between starch, protein, and fiber play a more critical role than ingredient proportion alone. Processing conditions such as steaming and baking may further modify fiber structure through partial solubilization or entrapment within the starch-protein network, thereby affecting its analytical recovery (Brennan et al., 2008). The higher fiber content observed in F2 may be attributed to a more balanced contribution of insoluble fiber, which is known to dominate cereal-based systems and largely determine total dietary fiber values (Brennan et al., 2008). These findings highlight that fiber functionality and quantification are governed by structural interactions within the food matrix rather than solely by compositional differences.

The total dietary fiber content was highest in flakes from formulation F3, reaching 4.91%, whereas formulation F1 exhibited the lowest value at 4.81%. Baker & Holden (2006) noted that over 75.3% of commercial cereals contain less than 5% dietary fiber, suggesting that flakes composed of white sweet potato and precooked pigeon peas have fiber contents comparable to typical commercial cereal products.

Dietary fiber intake plays a crucial role in preventing several NCDs, such as DM type 2 (Mazhar et al., 2023), colorectal cancer (Bestari et al., 2023), and it also reduces blood pressure (Jama et al., 2024). Moreover, findings from Reynolds et al. (2020) indicated that a daily intake of 19 g of dietary fiber can significantly lower HbA1c, fasting blood glucose, total cholesterol, LDL cholesterol, body weight, and body mass index.

Table 3 shows that formulation F2 had the highest resistant starch content (6.70%), while formulation F1 (30 g: 35 g) had the lowest (6.02%). This indicates that all formulations (F1, F2, and F3) can be categorized as high in resistant starch, with values ranging from 5% to 15%, consistent with the classification proposed by Birt et al. (2013) and supported by Afifah et al. (2020). Several factors influence the formation of resistant starch, including the starch-to-water ratio, autoclaving temperature, number of autoclaving-cooling cycles, the ratio of amylose to amylopectin, amylose chain length, and acid hydrolysis (Lasale et al., 2022). High levels of resistant starch are closely associated with elevated amylose content (Nakamura et al., 2016). This finding aligns with Yadav et al. (2009), who found that repeated heating and cooling during the precooking process increases amylose content, which consequently raises the resistant starch level in food products.

Increased amylose content contributes to higher resistant starch (RS), particularly RS type 3, through its structural and physicochemical properties. Upon thermal processing (e.g., autoclaving), starch granules undergo gelatinization, resulting in the leaching of amylose into the surrounding matrix. During subsequent cooling, these linear amylose chains reassociate via hydrogen bonding to form ordered double-helical crystalline structures, a process known as retrogradation. These retrograded structures are thermally stable and less susceptible to enzymatic hydrolysis, thereby increasing the fraction of resistant starch (Sajilata et al., 2006; Fuentes-Zaragoza et al., 2010).

Furthermore, higher amylose content enhances the formation of compact and dense starch networks, which restrict enzyme accessibility and reduce starch digestibility. Amylose can also interact with lipids to form amylose-lipid complexes (RS type 5), further contributing to resistance against enzymatic digestion (Birt et al., 2013). Therefore, starches with higher amylose content tend to produce greater amounts of resistant starch due to enhanced retrogradation, structural ordering, and reduced enzymatic susceptibility.

The increase in resistant starch observed in the composite flakes can be explained by the combined effects of autoclaving, cooling, and drying during processing. Autoclaving at high temperature and pressure (121 °C, 15 psi) promotes starch gelatinization, leading to the disruption of native granule structure and the leaching of amylose into the surrounding matrix. During the subsequent cooling phase, the solubilized amylose chains undergo retrogradation, reassociating into more ordered double-helical crystalline structures. These retrograded structures are less susceptible to enzymatic hydrolysis, thereby contributing to the formation of resistant starch, particularly RS type 3 (Sajilata et al., 2006; Fuentes-Zaragoza et al., 2010). Furthermore, the drying process stabilizes these recrystallized structures by reducing moisture content, which limits molecular mobility and preserves the resistant starch fraction. Therefore, the sequential processing conditions applied in this study (autoclaving-cooling-drying) are likely to enhance resistant starch formation through increased amylose availability and subsequent retrogradation, ultimately contributing to the lower digestibility and potential glycemic benefits of the composite flakes (Farooq & Yu, 2024). Kaur et al. (2024) mentioned that consumption of boiled basmati rice containing a total fiber of 3.5 g/100 g and resistant starch of 10.5 g/100 g for 28 days can reduce blood glucose by 29.7%, cholesterol by 37.9%, triglycerides by 31.3%, and LDL by 30.5% in diabetic rats. Therefore, flakes made from composite flour, white sweet potatoes, and precooked peas with fiber >4% and RS>6% have potential as a fiber food source to control blood glucose and prevent NCDs.

3.3 Reducing sugar

Table 4 mentioned that different ratios of white sweet potatoes and precooked peas have not affected the reducing sugar, with a P-value of 0.06. The mean of reducing sugar is between 1.40% and 1.44%.

Table 4
Reducing sugar.

Formulations	Reducing sugar (%) (mean±SD)	P-value
F1	1.44 ± 0.01a	0.06
F2	1.40 ± 0.02a
F3	1.43 ± 0.02ab

• The formulation of composite flour (wheat: sweet white potatoes, pre-cooked peas) was F1 (35:30:35), F2 (35:35:30) and F3 (35%:40%:25%). Different superscript letters indicate significant differences at a significance level of p < 0.05.

Based on the data presented in Table 4, F1 contained 1.44% reducing sugar, F2 contained 1.40%, and F3 contained 1.43%. These findings indicate that the different formulation ratios did not significantly affect reducing sugar content, as the variations among treatments were minimal and did not result in statistically significant differences. Reducing sugar content shows an inverse relationship with RS levels, where higher RS content is associated with lower reducing sugar in the product (Zheng et al., 2020). Reducing sugar is an indicator of the number of simple sugars in food that can be absorbed by the body. Reducing sugar is associated with the GI of a food product, where higher reducing sugar content typically corresponds to a higher GI (Najeeb et al., 2022). Therefore, foods consumed by individuals with diabetes should ideally have low reducing sugar content.

3.4 Metabolomic profile

The metabolic analysis is shown in Figure 1. 15 compounds had more than one VIP score. Most of the highly bioactive components come from the amino acid group. Several compounds appear to be positively associated with beneficial effects on DM based on their known biological roles in glucose metabolism, insulin sensitivity, and oxidative stress reduction, such as D-(–)-Quinic acid, 2,5-di-tert-butylhydroquinone (DTBHQ), L-Phenylalanine, 4-Guanidinobut also known as gamma-guanidinobutyrate or gamma amino acids, 12,13-diHOME and L-arginine (Zheng et al., 2025; Hossein et al., 2024).

Figure 1
Metabolic analysis.

Increasing the level of white sweet potatoes on flakes enhances the number of 2,5-di-tert-butylhydroquinone (DTBHQ), L-Phenylalanine, 4-Guanidinobut also known as gamma-guanidinobutyrate or gamma amino acids, 12,13-diHOME, L-Phenylalanine and L-arginine. An increased proportion of white sweet potatoes may contribute to higher glutamic acid levels in the flakes, which could enhance umami perception. Pan et al. (2023) reported that increased glutamic acid content was associated with enhanced umami taste in meat products; therefore, a similar mechanism may occur in these flakes. A study by Toyoizumi et al. (2021) reported that increasing thermal treatment temperatures significantly reduced γ-guanidinobutyrate levels in germinated brown rice. A similar mechanism may occur during the autoclaving process of peas, which could explain why increasing the substitution level of precooked peas did not lead to an increase in γ-guanidinobutyrate content.

The D-(-) Quinic acid was increasing in F2 and F3 and its bioactive exhibits strong antioxidant potential and may contribute to reducing oxidative stress, a key mechanism in preventing diabetic complications in liver, kidney and pancreas tissues of STZ-induced diabetic rats (Arya et al., 2014). Meanwhile, L-Phenylalanine has a good effect on gluconeogenesis and influences the secretion of insulin and glucagon, and decreases the duration of microvascular complications in patients with T2DM (Zheng et al., 2025). Furthermore, according to Macêdo et al. (2022), 12,13-diHOME has emerged as a potential therapeutic target for metabolic diseases. This oxylipin acts as a lipokine that regulates lipid and glucose metabolism by enhancing fatty acid uptake and oxidation in skeletal muscle and brown adipose tissue. The study demonstrated that elevated circulating levels of 12,13-diHOME are associated with improved insulin sensitivity and metabolic flexibility, suggesting its beneficial role in mitigating metabolic dysfunctions such as obesity and type 2 DM. However, this study has not yet been proven in an animal study; therefore, future studies need to continue potential of the F3 flakes to improve insulin resistance in T2DM.

3.5 Hardness

Table 5 shows that different proportions of white sweet potatoes and precooked peas affect hardness with a p-value of 0.01. The formula F1 that contained the highest proportion of precooked peas has a higher level of hardness compared to F2 and F3. However, control flakes that consist only of 100% wheat have the hardest hardness with a mean of 41.40 N.

Table 5
Hardness contents of flakes.

Formulations	Hardness (N) (mean±SD)	p-value
F1	21.08 ± 5.17a	0.01
F2	7.60 ± 0.15b
F3	12.39 ± 1.35b
Control	41.40 ± 3.20c

• Control referred to 100% wheat flour flakes. The formulation of composite flour (wheat: sweet white potatoes, pre-cooked peas) was F1 (35:30:35), F2 (35:35:30) and F3 (35:40:25).

The lower hardness was found in formulas F2 and F3. Brennan and Samyue (2004) reported that products with higher fat and lower fiber content tend to exhibit greater hardness due to reduced moisture retention and denser matrix formation. Meanwhile, higher content of dietary fiber and resistant starch present in precooked legumes and sweet potatoes in formula F2 and F3 might disrupt the starch matrix, absorb more water, and create a more open and less dense structure during the baking process. Such structural changes reduce the mechanical resistance of the product, resulting in lower hardness values (Zhang et al., 2023). Moreover, the highest amylose content in F3 might contribute to the reduction of hardness, since during the heating process, starch undergoes gelatinization, a phenomenon in which the hydrogen bonds maintaining the structural integrity of starch granules are disrupted. As water is absorbed, the granules swell significantly. When gelatinization occurs optimally, this swelling leads to a reduction in the hardness of the final product (Kyaw et al., 2001). This result is also in line with Kuchtová et al. (2018), who proved that the addition of both types of grape skin and seed in cookies increases total fiber but significantly decreases hardness, thickness and fracturability. Fitriani et al. (2018) also mentioned that increasing the banana flour that contained 19.5% of amylose has the highest hardness with a score of 5500 g/mm² in cereal made with broken rice and banana flour.

Besides amylose, hardness is also affected by protein and moisture content (Fitriani et al., 2018). This study proved that the more amylose and protein, the harder the banana and broken rice cereal will be. However, increasing temperature of drying temperature will reduce the moisture content and decrease the hardness. High protein can also interact with starch, forming complexes that restrict starch granule swelling, thereby contributing to increased product hardness in plant-based meal products (Auppasuk et al., 2025). The same theory may occur in this study. During thermal processing, water is absorbed into the starch matrix, initiating gelatinization. This process subsequently leads to moisture loss, particularly during the later stages of heating or drying, which enhances the hardness of the final product (Fitriani et al., 2018). This theory is in line with the result of this study that formula F3 significantly has lower hardness compared to F1 due to higher content of moisture, amylose and protein compared to F1.

3.6 Sensory acceptance

The analysis of sensory acceptance can be seen in Figure 2 and Table 6, while the picture of flakes is shown in Figure 3. The different ratios of white sweet potatoes and precooked peas affect the sensory acceptance of flakes. Increasing the number of substitutions of precooked pea flour decreased the aroma, texture and overall sensory acceptance of flakes with p-value 0.01 and <0.01. Formula 3 had the highest sensory acceptance based on aroma, texture and overall acceptance with a range score of 3.9 to 4.3. Increasing the proportion of white sweet potatoes elevated glutamic acid, enhancing umami taste, while also increasing phenylalanine associated with bitterness (Melis & Tomassini Barbarossa, 2017). Although Formula 3 (F3) had the lowest arginine content, its v-score remained high. As arginine contributes to sweetness perception (Melis & Tomassini Barbarossa, 2017), these results suggest that F3 exhibits a predominantly umami taste with minimal sweetness or bitterness.

Figure 2
Sensory acceptance of flakes.

Table 6
Sensory acceptance of flakes.

Formulations	Sensory (Mean ± SD)
	Color	Aroma	Taste	Texture	Overall
F1 (30 : 35)	3.60 ± 0.81	3.33 ± 0.80b	3.47 ± 1.01	3.00 ± 0.64b	3.50 ± 0.73b
F2 (35 : 30)	3.93 ± 0.98	3.63 ± 0.72ab	3.87 ± 0.77	3.80 ± 0.66a	3.90 ± 0.71a
F3 (40 : 25)	3.83 ± 0.83	3.97 ± 0.81a	3.97 ± 0.62	4.10 ± 0.61a	4.13 ± 0.63a
P-value	0.35	0.01	0.09	<0.01	<0.01

• The formulation of composite flour (wheat: sweet white potatoes, pre-cooked peas) was F1 (35:30:35), F2 (35:35:30) and F3 (35:40:25).

Figure 3
Picture of flakes.

An increased level of precooked peas decreased the aroma of flakes. This is in line with the study conducted by Ayuningtyas & Aan (2025) mentioned that increasing the substitution of red bean flour in cookie products decreases sensory acceptance. This might be because the use of red beans in large proportions can cause a bitter taste due to lipoxygenase activity. The same theory might occur in flakes with higher substitution of precooked peas (F1); therefore, the F3 flakes that contained the lowest amount of precooked peas were more acceptable in sensory terms compared to F1 and F2. The texture acceptability results showed that an increasing number of white sweet potatoes achieved the highest score at 4.10%. Hao et al. (2014) stated that fat content affects equilibrium moisture and molecular mobility of water in a food matrix, thereby altering the texture of food.

3.7 Glycemic Index (GI) and Glycemic Load (GL) analysis

The blood glucose response curve, as shown in Figure 4, demonstrates that the F1 and F3 flakes produced a lower postprandial blood glucose concentration compared to the control sample throughout the 120-minute observation period. The peak blood glucose level for the flakes group occurred at 15 minutes and was notably lower than that of the control, followed by a more gradual decline to baseline levels. This pattern indicates that the F1 and F3 flakes elicited a slower glucose release and absorption rate, reflecting improved carbohydrate quality and glycemic control potential.

Figure 4
Changes in blood glucose levels curve.

The GI and load GL are shown in Table 7. The study in healthy, normal BMI showed that the F1 flake has a GI of 32, while the F3 flake has a GI of 49, which both of them was classified as low, and the GL 5.49 and 8.79 were categorized as low GL (Handayani & Ayustaningwarno, 2014).

Table 7
GI and GL of F1 and F3 Flakes.

Formula	GI	Category	GL	Category
F1	32	Low	16.33	moderate
F3	49	Low	24.85	high

• The formulation of composite flour (wheat: white sweet potatoes, pre-cooked peas) was F1 (35:30:35) and F3 (35:40:25).

The present study highlights a clear relationship between dietary fiber composition and glycemic response of the developed flakes. The low glycemic index of the developed flakes may be attributed to their relatively high total dietary fiber, resistant starch, and amylose contents, as these components are known to slow starch digestion and attenuate postprandial glycemic responses (Soviana & Maenasari, 2019). Formulations F1 and F3, which exhibited relatively higher total dietary fiber (4.81% and 4.91%, respectively), were associated with lower GI values (32 and 49). However, despite both formulations being classified as low GI foods (GI <55), their GL values differed markedly, with F1 categorized as moderate (16.33) and F3 as high (24.85). This discrepancy highlights that GI alone does not fully reflect the overall glycemic impact of a food. GL combines the GI with the amount of carbohydrate consumed, and is thus primarily influenced by the available carbohydrate per serving. The difference in GL between F1 and F3 can be largely attributed to the variation in pea content. The lower inclusion of peas likely resulted in a higher proportion of rapidly digestible starch and increased available carbohydrate per serving, thereby elevating the GL (Foster-Powell et al., 2002; Augustin et al., 2015).

Furthermore, peas are rich in protein, dietary fiber, and resistant starch, all of which are known to modulate starch digestion. Dietary fiber, particularly soluble fiber, can increase intestinal viscosity and slow glucose absorption, while insoluble fiber may limit enzyme accessibility by physically entrapping starch within the food matrix (Elleuch et al., 2011). The interaction between starch, protein, and fiber can form a complex structure that limits enzymatic access. In F1, the higher content of peas likely enhanced matrix complexity, resulting in reduced starch hydrolysis. In contrast, F3 may have a more accessible starch structure, leading to faster digestion and higher GL (Singh et al., 2010).

Interestingly, although F3 had slightly higher total fiber than F1, its GI and GL were also higher. This suggests that total fiber alone is not the sole determinant of glycemic response. Other factors, such as total starch content, amylose structure, and the proportion of rapidly digestible starch, may have contributed to the higher glycemic response observed in F3. This indicates that the glycemic behavior of food is influenced by a complex interaction between compositional and structural factors rather than a single nutritional parameter.

Overall, the findings demonstrate that the combination of moderate dietary fiber, resistant starch, and appropriate processing techniques contributes to the development of flakes with low glycemic impact. These products can therefore be considered promising functional foods for glycemic control, particularly for individuals with or at risk of metabolic disorders such as type 2 diabetes.

From a nutritional perspective, foods with low GI and controlled GL have been shown to improve glycemic control, reduce insulin demand, and contribute to better lipid profiles (Jenkins et al., 2002; Brand-Miller & Buyken, 2012). A cross-sectional study conducted by Mayawati & Isnaeni (2017) mentioned that 68% of subjects with low intake of high GI tended to have higher blood glucose. Meanwhile, the moderate GL indicates that although the carbohydrate portion contributes moderately to total blood glucose response, it remains within a range that does not excessively elevate postprandial glucose (Augustin et al., 2015). Therefore, the flakes developed in this study, particularly F1, may serve as promising functional foods for the dietary management of metabolic disorders, including type 2 DM. By promoting a more stable glycemic response, such products may help prevent rapid glucose fluctuations associated with insulin resistance and disease progression (Gerontiti et al., 2024).

4 Conclusion

This study shows that different ratios of white sweet potatoes and precooked peas significantly affect the nutritional quality, physicochemical characteristics, sensory acceptance, and glycaemic response of composite flour flakes. Increasing precooked pea substitution enhanced protein, dietary fiber, resistant starch, and bioactive metabolites, while higher white sweet potato levels improved carbohydrate content, texture, sensory acceptance, and umami-related compounds. All formulations met Indonesian food standards for protein, fat, ash, and carbohydrate content, although moisture levels slightly exceeded the recommended limit. The F3 formulation (35:40:25) exhibited the most favourable balance, with high fiber (>4%), resistant starch (>6%), good sensory acceptance, lower hardness, and a low GI of 49. These findings suggest that white sweet potato-precooked pea flakes have strong potential as a functional cereal product for glycaemic control and the prevention of type 2 DM.

Acknowledgements

This research was supported by the Fundamental Research Grant from the Directorate of Research, Technology, and Community Service, Ministry of Education, Culture, Research, and Technology of the Republic of Indonesia, under master contract number 127/C3/DT.05.00/PL/2025. The authors express their sincere gratitude for the financial support and research facilitation provided.

Data Availability Statement

All data generated or analyzed in this study are included in this published article.

• Cite as:
  Mustikaningrum, F., Dewi, A., Zahrani, F., Kusuma, R., Wicaksana, A., Dayanti, A., Bestari, S., & Wang, W. (2026). Development of functional low glycemic index flakes using white sweet potatoes and precooked peas. Brazilian Journal of Food Technology, 29, e2025157. https://doi.org/10.1590/1981-6723.1572025
• Funding:
  Directorate of Research, Technology, and Community Service, Ministry of Education, Culture, Research, and Technology of the Republic of Indonesia, under master contract number 127/C3/DT.05.00/PL/2025.

References

• Afifah, N., Yuliana, N. D., Prangdimurti, E., & Meryandini, A. (2020). Resistant starch as a functional food for metabolic disorders: A review. Journal of Functional Foods, 64, 103667. https://doi.org/10.1016/j.jff.2019.103667
  » https://doi.org/10.1016/j.jff.2019.103667
• Akther, S., Alim, M. A., Badsha, M. R., Matin, A., Ahmad, M. I., & Hoque, S. (2020). Formulation and quality evaluation of instant mango drink powder. Food Research, 4(4), 1287-1296. https://doi.org/10.26656/fr.2017.4(4).077
  » https://doi.org/10.26656/fr.2017.4(4).077
• Arya, A., Jamil Al-Obaidi, M. M., Shahid, N., Bin Noordin, M. I., Looi, C. Y., Wong, W. F., Khaing, S. L., & Mustafa, M. R. (2014). Synergistic effect of quercetin and quinic acid by alleviating structural degeneration in the liver, kidney and pancreas tissues of STZ-induced diabetic rats: A mechanistic study. Food and Chemical Toxicology: An International Journal Published for the British Industrial Biological Research Association, 71, 183-196. PMid:24953551. https://doi.org/10.1016/j.fct.2014.06.010
  » https://doi.org/10.1016/j.fct.2014.06.010
• Association of Official Analytical Chemists – AOAC. (1995). Official methods of analysis of the Association of Official Analytical Chemists Gaithersburg: AOAC.
• Augustin, L. S. A., Kendall, C. W. C., Jenkins, D. J. A., Willett, W. C. A., Astrup, A. W., Barclay, I., Björck, J. C., Brand-Miller, J. C., Brighenti, F., Buyken, A. E., Ceriello, A., La Vecchia, C., Livesey, G., Liu, S., Riccardi, G., Rizkalla, S. W., Sievenpiper, J. L., Trichopoulou, A., Wolever, T. M. S., Baer-Sinnott, S., & Poli, A. (2015). Glycemic index, glycemic load and glycemic response: An International Scientific Consensus Summit from the International Carbohydrate Quality Consortium (ICQC). Nutrition, Metabolism, and Cardiovascular Diseases: NMCD, 25(9), 795-815. PMid:26160327. https://doi.org/10.1016/j.numecd.2015.05.005
  » https://doi.org/10.1016/j.numecd.2015.05.005
• Auppasuk, P., Chaisuwan, B., Ploypetchara, T., Butseekhot, S., & Sorndech, W. (2025). Influence of elevated protein content in textured vegetable protein on the textural characteristics and sensory attributes of plant-based meat alternatives. Applied Food Research, 5(2), 101192. https://doi.org/10.1016/j.afres.2025.101192
  » https://doi.org/10.1016/j.afres.2025.101192
• Ayuningtyas, N. T., & Aan, S. (2025). Sifat kimia dan organoleptik cookies bebas gluten bebas kasein berbasis tepung komposit mocaf dan kacang merah. Journal of Multidiciplinary Research and Development, 7(4), 2865-2876.
• Baker, D., & Holden, J. M. (2006). Fiber in breakfast cereals. Journal of Food Science, 46(2), 396-398. https://doi.org/10.1111/j.1365-2621.1981.tb04869.x
  » https://doi.org/10.1111/j.1365-2621.1981.tb04869.x
• Behall, K., & Hallfrisch, J. (2002). Plasma glucose and insulin reduction after consumption of breads varying in amylose content. European Journal of Clinical Nutrition, 56(9), 913-920. PMid:12209381. https://doi.org/10.1038/sj.ejcn.1601411
  » https://doi.org/10.1038/sj.ejcn.1601411
• Bestari, S. A., Fulyani, F., Kusuma, R. J., & Lestari, E. (2023). A diet high in protein and fiber changes the gut microbiota of the colorectal cancer rat model. Food Research, 7(3), 221-226. https://doi.org/10.26656/fr.2017.7(3).575
  » https://doi.org/10.26656/fr.2017.7(3).575
• Birt, D. F., Boylston, T., Hendrich, S., Jane, J. L., Hollis, J. L. L., McClelland, J., Moore, S., Phillips, G. J., Rowling, M., Schalinske, K., Scott, M. P., & Whitley, E. M. (2013). Resistant starch: Promise for improving human health. Advances in Nutrition, 4(6), 587-601. PMid:24228189. https://doi.org/10.3945/an.113.004325
  » https://doi.org/10.3945/an.113.004325
• Brand-Miller, J., & Buyken, A. E. (2012). The glycemic index issue. Current Opinion in Lipidology, 23(1), 62-67. PMid:22157060. https://doi.org/10.1097/MOL.0b013e32834ec705
  » https://doi.org/10.1097/MOL.0b013e32834ec705
• Brennan, S. C., & Samyue, E. (2004). Evaluation of starch degradation and textural characteristics of dietary fiber enriched biscuits. International Journal of Food Properties, 7(3), 647-657. https://doi.org/10.1081/JFP-200033070
  » https://doi.org/10.1081/JFP-200033070
• Brennan, M. A., Monro, J. A., & Brennan, C. S. (2008). Effect of inclusion of soluble and insoluble fibres into extruded breakfast cereal products made with reverse screw configuration. International Journal of Food Science & Technology, 43(12), 2278-2288. https://doi.org/10.1111/j.1365-2621.2008.01867.x
  » https://doi.org/10.1111/j.1365-2621.2008.01867.x
• Chen, F., Song, Y., Dong, M., Xie, Y., Qiao, Z., Zhu, Y., Cao, P., Yu, Y., & Yuan, C. (2025). A high-throughput method for determining the amylose content of rice. PLoS One, 20(11), e0336600. PMid:41289244. https://doi.org/10.1371/journal.pone.0336600
  » https://doi.org/10.1371/journal.pone.0336600
• Correia, B. S. B., Sørensen, B. E., Aaslyng, D. A., & Bertram, C. H. (2025). Metabolome of different cultivars of peas, lentils, faba beans and lupins: An 1H NMR spectroscopic exploration of their sensory attributes and potential biofunctionality. Food Chemistry, 477, 143579. PMid:40020623. https://doi.org/10.1016/j.foodchem.2025.143579
  » https://doi.org/10.1016/j.foodchem.2025.143579
• Dwi, A. R., & Rohmawati, N. (2019). Glycemic index, nutrient content, and the acceptance Capability Sweet Potato (Ipomoea batatas) pudding with addition of red dragon fruit (Hylocereus polyrhizus). Amerta Nutrition, 3(2), 107-113.
• Elleuch, M., Bedigian, D., Roiseux, O., Besbes, S., Blecker, C., & Attia, H. (2011). Dietary fibre and fibre-rich by-products. Food Chemistry, 124(2), 411-421. https://doi.org/10.1016/j.foodchem.2010.06.077
  » https://doi.org/10.1016/j.foodchem.2010.06.077
• Farooq, M. A., & Yu, J. (2024). Recent advances in physical processing techniques to enhance the resistant starch content in foods: A review. Foods, 13(17), 2770. PMid:39272535. https://doi.org/10.3390/foods13172770
  » https://doi.org/10.3390/foods13172770
• Fitriani, V., Permana, L., & Setiaboma, W. (2018). Chemical and physical characterisation of cereal flakes formulated with broken rice and banana flour. IOP Conference Series. Earth and Environmental Science, 258, 012003. https://doi.org/10.1088/1755-1315/258/1/012003
  » https://doi.org/10.1088/1755-1315/258/1/012003
• Food and Agriculture Organization – FAO. (1997). Codex Alimentarius: Guidelines for use of nutrition and health claims number CAC/GL 23-1997. Rome: FAO. Retrieved in 2025, December 17, from https://www.fao.org/ag/humannutrition/32443-04352e8311b857c57caf 5fc4c5c4a4cd.pdf
  » https://www.fao.org/ag/humannutrition/32443-04352e8311b857c57caf
• Foster-Powell, K., Holt, S. H., & Brand-Miller, J. C. (2002). International table of glycemic index and glycemic load values. The American Journal of Clinical Nutrition, 76(1), 5-56. PMid:12081815. https://doi.org/10.1093/ajcn/76.1.5
  » https://doi.org/10.1093/ajcn/76.1.5
• Fuentes-Zaragoza, E., Riquelme-Navarrete, M. J., Sánchez-Zapata, E., & Pérez-Álvarez, J. A. (2010). Resistant starch as functional ingredient: A review. Food Research International, 43(4), 931-942. https://doi.org/10.1016/j.foodres.2010.02.004
  » https://doi.org/10.1016/j.foodres.2010.02.004
• Gerontiti, E., Shalit, A., Stefanaki, K., Kazakou, P., Karagiannakis, D. S., Peppa, M., Psaltopoulou, T., & Paschou, S. A. (2024). The role of low glycemic index and load diets in medical nutrition therapy for type 2 diabetes: An update. Hormones: International Journal of Endocrinology and Metabolism, 23(4), 655-665. PMid:38750304. https://doi.org/10.1007/s42000-024-00566-7
  » https://doi.org/10.1007/s42000-024-00566-7
• Handayani, L., & Ayustaningwarno, F. (2014). Indeks glikemik dan beban glikemik vegetable leather brokoli (Brassica oleracea var Italica) dengan substitusi inulin. Journal of Nutrition College, 3(4), 783-790. https://doi.org/10.14710/jnc.v3i4.6881
  » https://doi.org/10.14710/jnc.v3i4.6881
• Hao, F., Lu, L., & Ge, C. (2014). Effect of fat content on water sorption properties of biscuits studied by nuclear magnetic resonance. Journal of Food and Nutrition Research, 2(11), 814-818. https://doi.org/10.12691/jfnr-2-11-9
  » https://doi.org/10.12691/jfnr-2-11-9
• Hossein, R., Mohammad, S. R., Mohmmadreza, S., & Nepton, S. (2024). Gamma-aminobutyric acid attenuates insulin resistance in type 2 diabetic patients and reduces the risk of insulin resistance in their offspring. Biomedicine & Pharmacotherapy = Biomédecine & Pharmacothérapie, 3, 111440. https://doi.org/10.1016/j.biopha.2021.111440
  » https://doi.org/10.1016/j.biopha.2021.111440
• Howlett, J., & Ashwell, M. (2008). Glycemic response and health: Summary of a workshop. The American Journal of Clinical Nutrition, 87(1), 212S-216S. PMid:18175760. https://doi.org/10.1093/ajcn/87.1.212S
  » https://doi.org/10.1093/ajcn/87.1.212S
• Islam, S. (2024). Sweetpotatoes [Ipomoea batatas(L.) lam]: The super food of the Next Century? An intensive review on their potential as a sustainable and versatile food source for future generations. CYTA: Journal of Food, 22(1), 2397553. https://doi.org/10.1080/19476337.2024.2397553
  » https://doi.org/10.1080/19476337.2024.2397553
• Jama, H. A., Snelson, M., Schutte, A. E., Muir, J., & Marques, F. Z. (2024). Recommendations for the use of dietary fiber to improve blood pressure control. Hypertension, 81(7), 1450-1459. PMid:38586958. https://doi.org/10.1161/HYPERTENSIONAHA.123.22575
  » https://doi.org/10.1161/HYPERTENSIONAHA.123.22575
• Jenkins, D. J., Kendall, C. W., Augustin, L. S., Franceschi, S., Hamidi, M., Marchie, A., Jenkins, A. L., & Axelsen, M. (2002). Glycemic index: Overview of implications in health and disease. The American Journal of Clinical Nutrition, 76(1), 266S-273S. PMid:12081850. https://doi.org/10.1093/ajcn/76.1.266S
  » https://doi.org/10.1093/ajcn/76.1.266S
• Kaur, P., Kaur, H., Aggarwal, R., Bains, K., Mahal, A. K., Singla, L. D., & Gupta, K. (2024). Analysing the impact of resistant starch formation in basmati rice products: Exploring associations with blood glucose and lipid profiles across various cooking and storage conditions in vivo Foods, 13(11), 1669. PMid:38890898. https://doi.org/10.3390/foods13111669
  » https://doi.org/10.3390/foods13111669
• Kuchtová, V., Kohajdová, Z., Karovičová, J., & Lauková, M. (2018). Physical, textural and sensory properties of cookies incorporated with grape skin and seed preparations. Polish Journal of Food and Nutrition Sciences, 68(4), 309-317. https://doi.org/10.2478/pjfns-2018-0004
  » https://doi.org/10.2478/pjfns-2018-0004
• Kyaw, Z. Y., Cheow, C. S., Yu, S. Y., & Dzulkifly, M. (2001). The effect of pressure cooking on the microstructure and expansion of fish crackers (keropak). Journal of Food Quality, 24(3), 181-194. https://doi.org/10.1111/j.1745-4557.2001.tb00601.x
  » https://doi.org/10.1111/j.1745-4557.2001.tb00601.x
• Lao, Y., Ye, Q., Wang, Y., & Selomulya, C. (2024). Modulating digestibility and mitigating beany flavour of pea protein. Food Reviews International, 40(9), 3129-3158. https://doi.org/10.1080/87559129.2024.2346329
  » https://doi.org/10.1080/87559129.2024.2346329
• Lasale, V., Nurfadilla, N., & Anwar, E. (2022). The impact of autoclaving-cooling cycles on resistant starch formation in tuber-based foods. Food Hydrocolloids, 128, 107603. https://doi.org/10.1016/j.foodhyd.2022.107603
  » https://doi.org/10.1016/j.foodhyd.2022.107603
• Low, W. Y., Lee, Y. K., & Samy, A. L. (2015). Non-communicable diseases in the Asia-Pacific region: Prevalence, risk factors and community-based prevention. International Journal of Occupational Medicine and Environmental Health, 28(1), 20-26. PMid:26159943. https://doi.org/10.2478/s13382-014-0326-0
  » https://doi.org/10.2478/s13382-014-0326-0
• Macêdo, A. P. A., Muñoz, V. R., Cintra, D. E., & Pauli, J. R. (2022). 12,13-diHOME as a new therapeutic target for metabolic diseases. Life Sciences, 290, 120229. PMid:34914931. https://doi.org/10.1016/j.lfs.2021.120229
  » https://doi.org/10.1016/j.lfs.2021.120229
• Martínez, Y., Li, X., Liu, G., Bin, P., Yan, W., Más, D., Valdivié, M., Hu, C. A., Ren, W., & Yin, Y. (2017). The role of methionine on metabolism, oxidative stress, and diseases. Amino Acids, 49(12), 2091-2098. PMid:28929442. https://doi.org/10.1007/s00726-017-2494-2
  » https://doi.org/10.1007/s00726-017-2494-2
• Mayawati, H., & Isnaeni, N. F. (2017). Hubungan asupan makanan indeks glikemik tinggi dan aktivitas fisik dengan kadar glukosa darah pada pasien diabetes mellitus tipe II Rawat Jalan di RSUD Karanganyar. Jurnal Kesehatan, 10(1), 75-83. https://doi.org/10.23917/jk.v10i1.5495
  » https://doi.org/10.23917/jk.v10i1.5495
• Mazhar, M., Zhu, Y., & Qin, L. (2023). The interplay of dietary fibers and intestinal microbiota affects type 2 diabetes by generating short-chain fatty acids. Foods, 12(5), 1023. PMid:36900540. https://doi.org/10.3390/foods12051023
  » https://doi.org/10.3390/foods12051023
• Melis, M., & Tomassini Barbarossa, I. (2017). Taste perception of sweet, sour, salty, bitter, and umami and changes due to L-arginine supplementation, as a function of genetic ability to taste 6-n-propylthiouracil. Nutrients, 9(6), 541. PMid:28587069. https://doi.org/10.3390/nu9060541
  » https://doi.org/10.3390/nu9060541
• Mustikaningrum, F., Marsono, Y., & Murdiati, A. (2013). Pengaruh pratanak kacang kapri (Pisum sativum L) terhadap kadar pati dan sifat hipoglikemiknya pada tikus diabetik induksi alloksan (Master's thesis). Universitas Gadjah Mada, Yogyakarta.
• Najeeb, I. M. A., Norhayati, M. K., Zaharah, H., Isa, N. S. M., Nadrah, M. R. N., & Yusof, H. M. (2022). Physicochemical properties, sensory acceptance and glycaemic index of processed stingless bee honey and processed honeybee honey. Food Research, 6(6), 103-110. https://doi.org/10.26656/fr.2017.6(6).745
  » https://doi.org/10.26656/fr.2017.6(6).745
• Nakamura, Y., Fujita, N., & Satoh, H. (2016). Starch biosynthesis and its regulation in cereals. Plant Biotechnology Journal, 14(3), 447-455. https://doi.org/10.1111/pbi.12370
  » https://doi.org/10.1111/pbi.12370
• National Standardization Agency. (2000). SNI 01-2886:2000: Extruded snacks standard Indonesia.
• Pan, L., Zhang, C.-J., Bai, Z., Liu, Y.-Y., Zhang, Y., Tian, W., Zhou, Y., Zhou, Y., Liao, A., Hou, Y., Yu, G., Hui, M., & Huang, J. (2023). Effects of different strains fermentation on nutritional functional components and flavor compounds of sweet potato slurry. Frontiers in Nutrition, 10, 1241580. PMid:37693241. https://doi.org/10.3389/fnut.2023.1241580
  » https://doi.org/10.3389/fnut.2023.1241580
• Paula, C. D., Pastrana-Puche, Y. I., Viloria-Benítez, K. M., Rubio-Arrieta, J. A., Simanca-Sotelo, M., Álvarez-Badel, B., & Avilez-Montes, Y. (2021). Physicochemical and sensory evaluation of sweet potato (Ipomoea batatasL.) restructured products produced in the Sinu Valley, Colombia. Heliyon, 7(8), e07691. PMid:34401577. https://doi.org/10.1016/j.heliyon.2021.e07691
  » https://doi.org/10.1016/j.heliyon.2021.e07691
• Peres, M., Costa, H. S., Silva, M. A., & Albuquerque, T. G. (2023). The Health effects of low glycemic index and low glycemic load interventions on prediabetes and type 2 diabetes mellitus: A literature review of RCTs. Nutrients, 15(24), 5060. PMid:38140319. https://doi.org/10.3390/nu15245060
  » https://doi.org/10.3390/nu15245060
• Reynolds, A. N., Akerman, A. P., & Mann, J. (2020). Dietary fibre and whole grains in diabetes management: Systematic review and meta-analyses. PLoS Medicine, 17(3), e1003053. PMid:32142510. https://doi.org/10.1371/journal.pmed.1003053
  » https://doi.org/10.1371/journal.pmed.1003053
• Sajilata, M. G., Singhal, R. S., & Kulkarni, P. R. (2006). Ressistant starch: A review. Comprehensive Reviews in Food Science and Food Safety, 5(1), 1-17. PMid:33412740. https://doi.org/10.1111/j.1541-4337.2006.tb00076.x
  » https://doi.org/10.1111/j.1541-4337.2006.tb00076.x
• Saman, W. R., Yuliasih, I., & Sugiarto, (2020). Physicochemical characteristics and functional properties of white sweet potato starch. International Journal of Engineering and Management Research, 9(3), 53-57. https://doi.org/10.31033/ijemr.9.3.7
  » https://doi.org/10.31033/ijemr.9.3.7
• Singh, J., Dartois, A., & Kaur, L. (2010). Starch digestibility in food matrix. Trends in Food Science & Technology, 21(4), 168-180. https://doi.org/10.1016/j.tifs.2009.12.001
  » https://doi.org/10.1016/j.tifs.2009.12.001
• Soviana, E., & Maenasari, D. (2019). Asupan serat, beban glikemik dan kadar glukosa darah pada pasien diabetes melitus tipe 2. Jurnal Kesehatan, 12(1), 19-29. https://doi.org/10.23917/jk.v12i1.8936
  » https://doi.org/10.23917/jk.v12i1.8936
• Suryaningrum, T., & Rustanti, N. (2016). Pengaruh Perbandingan Tepung Labu Kuning (Cucurbita moschata) Dan Tepung Mocaf Terhadap Kadar Kadar Pati, Nilai Indeks Glikemik (IG), Beban Glikemik (BG), Dan Tingkat Kesukaan Pada Flakes Kumo. Journal of Nutrition College, 5(4), 360. Retrieved in 2025, December 17, from http://ejournal-s1.undip.ac.id/index.php/jnc
  » http://ejournal-s1.undip.ac.id/index.php/jnc
• Teow, C. C., Truong, V.-D., McFeeters, R. F., Thompson, R. L., Pecota, K. V., & Yencho, G. C. (2007). Antioxidant activities, phenolic and β-carotene contents of sweet potato genotypes with varying flesh colours. Food Chemistry, 103(3), 829-838. https://doi.org/10.1016/j.foodchem.2006.09.033
  » https://doi.org/10.1016/j.foodchem.2006.09.033
• Totosaus, A., Montejano, J. G., Salazar, J. A., & Guerrero, I. (2002). A review of physical and chemical protein-gel induction. International Journal of Food Science & Technology, 37(6), 589-601. https://doi.org/10.1046/j.1365-2621.2002.00623.x
  » https://doi.org/10.1046/j.1365-2621.2002.00623.x
• Toyoizumi, T., Kosugi, T., Toyama, Y., & Nakajima, T. (2021). Effects of high-temperature cooking on the gamma-aminobutyric acid content and antioxidant capacity of germinated brown rice (Oryza sativa L.). CYTA: Journal of Food, 19(1), 360-369. https://doi.org/10.1080/19476337.2021.1905721
  » https://doi.org/10.1080/19476337.2021.1905721
• Vannini, M., Marchese, P., Sisti, L., Saccani, A., Mu, T., Sun, H., & Celli, T. (2021). Integrated efforts for the valorisation of sweet potato by-products within a circular economy concept: Biocomposites for packaging applications close the loop. Polymers, 13(7), 1048. PMid:33801582. https://doi.org/10.3390/polym13071048
  » https://doi.org/10.3390/polym13071048
• Wan, X., Wang, X., & Xiao, Q. (2024). Comparative metabolomic analysis of the nutrient composition of different varieties of sweet potato. Molecules, 29(22), 5395. PMid:39598785. https://doi.org/10.3390/molecules29225395
  » https://doi.org/10.3390/molecules29225395
• Winham, D. M., Camacho-Arriola, M., Glick, A. A., Hall, C. A., & Shelley, M. C. (2025). Pea and lentil flours increase postprandial glycemic response in adults with type 2 diabetes and metabolic syndrome. Foods, 14(11), 1933. PMid:40509460. https://doi.org/10.3390/foods14111933
  » https://doi.org/10.3390/foods14111933
• Yadav, B., Sharma, A., & Yadav, R. (2009). Studies on effect of multiple heating/cooling cycles on the resistant starch formation in cereals, legumes and tubers. International Journal of Food Sciences and Nutrition, 60(sup4, Suppl.4), 258-272. PMid:19562607. https://doi.org/10.1080/09637480902970975
  » https://doi.org/10.1080/09637480902970975
• Zhang, Y., Li, F., Huang, K., Li, S., Cao, H., Xie, J., & Guan, X. (2023). Structural changes of starch under different milling degrees affect the cooking and textural properties of rice. Food Chemistry: X, 17, 100627. PMid:36974186. https://doi.org/10.1016/j.fochx.2023.100627
  » https://doi.org/10.1016/j.fochx.2023.100627
• Zheng, J., Zhang, W., Miao, Y. Y., Li, X. R., Luo, W. M., Yang, X. L., Fang, Z. Z., & Zhang, Q. (2025). Interactive effect of phenylalanine with duration of diabetes on the risk of small vessel disease in Chinese patients with T2DM. Frontiers in Endocrinology, 15, 1472967. PMid:39829954. https://doi.org/10.3389/fendo.2024.1472967
  » https://doi.org/10.3389/fendo.2024.1472967
• Zheng, Y., Wei, Z., Zhang, R., Deng, Y., Tang, X., Zhang, Y., Liu, G., Liu, L., Wang, J., Liao, N., & Zhang, M. (2020). Optimization of the autoclave preparation process for improving resistant starch content in rice grains. Food Science & Nutrition, 8(5), 2383-2394. PMid:32405395. https://doi.org/10.1002/fsn3.1528
  » https://doi.org/10.1002/fsn3.1528

Edited by

• Associate Editor:
  Fabiana Andrea B. Galland.

Publication Dates

• Publication in this collection
  17 Aug 2026
• Date of issue
  2026

History

• Received
  17 Dec 2025
• Accepted
  29 May 2026