Study on the physicochemical and sensory profile of pliek-u: A traditional dried fermented coconut endosperm from Aceh, Indonesia

1.   Introduction

Coconut (Cocos nucifera) is a well-known plant used in Indian and African systems of medicine. The coconut fruit consists of the following layers: exocarp, mesocarp, endocarp, endosperm and embryo ^[1]. In Acehnese culture, we have an old tradition of processing coconut endosperm to produce pliek-u and coconut oil ^[2]. Pliek-u is dried fermented coconut endosperm. Presently, Acehnese society uses pliek-u as a spice in a special soup called kuah pliek, in a fruit salad called rujak for certain occasions, as well as in sauce ^[3]. However, the consumption rate of pliek-u tends to be low since the community does not know about its functional properties.

In other parts of the world, coconut oil is commonly produced through the fermentation of coconut milk ^[4,5,6], as well as by drying coconut kernels ^[7,8]. The tradition of processing coconut endosperm through fermentation, which is called pliek-u processing, only exists in Aceh. In batch processing, pliek-u processors can not only produce pliek-u, but also virgin coconut oil and crude coconut oil. Since it is made through a highly efficient process, we want to conserve the tradition of pliek-u processing. The conservation of this tradition can be used as a tool to improve the economy and public health ^[9].

There are about six species of fungi (Microascus sp., Sordaria sp., Curvularia sp., Trichurus sp., Acremonium sp. and Gonytrichum sp.) and approximately three species of gram-positive bacteria identified in pliek-u processing ^[3]. Another study has also shown nine isolates of fungi consisting of Aspergillus sp., Penicillium citrinum, Thielaviopsis paradoxa and Geotrichum candidum. Pliek-u contains lactic acid bacteria, and it can inhibit the growth of Escherichia coli and Staphylococcus aureus ^[10]. The gas chromatography–mass spectrometry (GC-MS) analysis had shown that pliek-u contained antimicrobial compounds as a result of fermentation ^[11]. Recently, the bioinformatic approach has shown the potential of the fatty acid in pliek-u as a component for atopic dermatitis treatment ^[12]. The antibacterial activity of pliek-u has been reported to be at a concentration of 5% ^[13] and at concentration of 4% against S. aureus ^[14]. The crude coconut oil also contains oleic acid, which can be used as a component for atopic dermatitis treatment ^[15]; and, its linoleic acid can be used against SARS-CoV-2 ^[16]. The use of pliek-u in feed had also improved the growth of African catfish significantly ^[17]. In contrast, the incidence of mycotoxins also degrades the quality of pliek-u after the infestation of Aspergillus flavus ^[18]. These findings motivate us to determine both the physicochemical and sensory quality of pliek-u.

Pliek-u processing consists of several steps, e.g., grating, fermentation, drying and pressing, and it takes a long time, i.e., about 13 to 16 days. The people who process pliek-u use traditional technologies that result in low productivity and quality. There are limited studies about improving the quality of pliek-u. The optimization of pliek-u quality has been achieved through the use of a kinetic dryer ^[19] and a Hohenheim solar tunnel dryer ^[9]. The quality of pliek-u could be determined by using near infrared spectroscopy ^[20]. The shelf life of pliek-u has been determined through the construction of an Arrhenius model ^[2], and the use of glass packaging can extend its shelf life ^[21] However, the standard quality of pliek-u has not been determined. Therefore, the objective of this study was to determine the physicochemical and sensory quality of pliek-u produced by random processors in Aceh. The results of this study are expected to help improve the understanding of pliek-u quality. Last but not least, we hope that this information will contribute to the standardization of pliek-u processing and prevention of mycotoxin contamination, thereby conserving this tradition and increasing its consumption rate.

2.   Materials and methods

2.1.   Research procedure

A survey study was conducted in five local markets, namely, Rukoh, Ulee Kareng, Al Mahirah, Kampong Baro and Lambaro, to get random samples of pliek-u from various processors in Aceh, Indonesia. We conducted a survey in these local markets due to their locations in two districts i.e., Banda Aceh and Aceh Besar, as well as their large size, which made complete products available; thus, we had a high possibility of finding pliek-u products. A set of questionnaires built with KoboToolbox (https://kf.kobotoolbox.org/) was used to collect the samples of pliek-u. There were six questions that focused on the names of sellers, name of the market, origin of the pliek-u, price per kilogram, sample number and location of the market by coordinates. In these local markets, we looked for pliek-u, retrieved some information from the sellers according to the questionnaires and then bought the products to be later used as samples.

Another set of questionnaires built using KoboToolbox was also used for an organoleptic test. An organoleptic test was conducted using the hedonic scale. A hedonic score of 5 was applied, where 1 refers to extremely dislike, 2 refers to dislike, 3 refers to neither like nor dislike, 4 refers to like and 5 refers to extremely like. All collected samples were evaluated by experts on the organoleptic properties of aroma, color, texture and taste. In addition, there were 11 questions about the characteristics of pliek-u, recipes, pliek-u conservation and the consumption rate. We chose the three experts on the basis of their expertise as related to the products, i.e., their skill and experiences in the production of pliek-u, as well as their experience in using pliek-u in their recipes. To find these experts, we used snowball sampling.

2.2.   Research parameters

The samples of pliek-u collected for the survey were stored in a plastic bag and sent to a laboratory for quality analysis. The parameters observed were the moisture, ash, fat, protein, fiber and carbohydrate contents, and the pH value and color. All parameters were analyzed by following the Indonesian National Standard (SNI) No. 01-2891-1992 in duplicate; the moisture content was analyzed by using the drying oven method at 105 ℃; the ash content was analyzed by conducting gravimetric analysis following implementation of the dry ashing method at 550 ℃; the fat content was analyzed via a direct extraction method; the protein content was analyzed by using the Kjeldahl method; and the fiber content was analyzed by using a gravimetric method. The carbohydrate content was calculated by using the pass difference method (it was by subtracting the measured water, protein, fat and ash from the total percentage) as it was used in the United States and Canada. The pH value was measured by using a pH meter (HM PH-80). The color of the pliek-u was determined by using a digital colorimeter (LS171) with a scale of Red (R), Green (G) and Blue (B).

For the three best samples with the highest organoleptic scores, the volatile compounds were analyzed by conducting the GC-MS method at the Laboratory of Science and Technology Faculty at State Islamic University Ar-Raniry. It is an internationally accepted analytical comparative technique whereby the compounds are separated in the gas phase and a characteristic pattern of separate compounds is obtained ^[22]. PerkinElmer apparatuses were used ^[23]. A PerkinElmer Turbomatrix HS 40 Headspace Trap automatic sampler with trap enrichment was used to extract the volatile compounds. Volatile compounds were analyzed using a PerkinElmer Clarus 680GC with an SQ8 mass spectrometer. They were separated on a capillary column with dimensions of 30 m × 0.25 mm, with helium as the mobile phase. The initial temperature of the oven, 40 ℃, was maintained for 3 min. An increase in temperature was applied at rate of 3 ℃/min up to 115 ℃ and held for 10 min, and then at 2 ℃/min up to 140 ℃ and kept for 5 min. The injector temperature was 210 ℃, and the transfer temperature was 150 ℃. The mass spectrometer was operated in the scan mode within a range of 45–500 AMU.

2.3.   Data analysis

Data were analyzed by using Microsoft Excel, including the statistical descriptives of mean value and standard deviation, and by building a map of the processor regions. Data were also analyzed by performing principal component analysis (PCA) using IBM SPSS Statistic 20.

3.   Results

3.1.   The distribution of pliek-u processors in Aceh

Aceh Province is located at the northwestern end of Sumatra Island, Indonesia. According to our survey, we collected 12 samples of pliek-u that were produced in four districts, i.e., Bireuen (eight samples), Banda Aceh (two samples), Aceh Besar (one sample) and Aceh Timur (one sample). The results showed that the central producer of pliek-u was the Bireuen subdistrict, which supplied about 67% of the pliek-u in the market. Aceh province has about 23 subdistricts; however, in other areas, a pliek-u processor was rarely found. Therefore, the conservation of this product should be initiated immediately.

From the survey, we also collected the information that the price of pliek-u in the local markets ranged from IDR 30,000 (USD 1.98) to IDR 50,000 (USD 3.3). Pliek-u was sold traditionally without labeling and packaging. Figure 1 shows the distribution of pliek-u processors in Aceh; the map was built by using Microsoft Excel.

Figure 1.  The distribution of pliek-u processors in Aceh, Sumatra Island.

DownLoad: Full-Size Img PowerPoint

3.2.   The proximate analysis of pliek-u

Proximate analysis was conducted for all samples collected from the survey, as can be seen in Table 1. The moisture content of pliek-u ranged from 8% to 19%, the ash content ranged from 2% to 8%, the fat content ranged from 27% to 50%, the protein content ranged from 17% to 24%, the fiber content ranged from 2% to 5%, the carbohydrate content ranged from 13% to 29% and the pH value ranged from 3.97 to 5.04. Two samples were found to be moldy after a week of storage, i.e., sample numbers 2 and 10. High standard deviations were observed for fat content, carbohydrate content and moisture content. The variation in the physicochemical quality of the pliek-u was caused by the unstandardized quality of processing among the processors.

Table 1.  Results of proximate analysis of pliek-u.

Sample number	Origin	Moisture (%)	Ash (%)	Fat (%)	Protein (%)	Fiber (%)	Carbohydrate (%)	pH
1	Bireuen	15.46	5.71	33.28	20.01	3.26	25.54	4.05
2	Bireuen	15.84	4.37	31.81	18.24	3.48	29.74	4.85
3	Bireuen	13.34	5.19	39.92	23.59	2.73	17.97	4.27
4	Banda Aceh	18.16	5.77	38.76	19.18	2.87	18.12	4.32
5	Bireuen	15.02	7.81	27.71	20.93	2.31	28.54	3.99
6	Bireuen	12.09	5.95	35.23	18.27	4.71	28.46	3.97
7	Bireuen	13.80	4.86	49.30	17.40	4.36	14.63	4.47
8	Bireuen	16.63	6.88	34.03	18.35	2.20	24.11	4.09
9	Aceh Timur	8.30	4.66	51.44	19.98	4.69	15.63	4.29
10	Banda Aceh	19.32	5.12	29.06	20.01	2.17	26.49	5.04
11	Bireuen	17.76	2.34	34.99	20.89	3.18	24.03	4.04
12	Aceh Besar	12.38	6.28	45.99	21.76	2.35	13.59	4.25
Mean		14.84	5.41	37.63	19.88	3.19	22.24	4.30
Standard deviation		3.07	1.37	7.71	1.75	0.95	5.89	0.34
Note: Sample numbesr 2 and 10 were moldy after a week of storage

---

 | Show Table

DownLoad: CSV

3.3.   Organoleptic test of pliek-u

Three experts were interviewed to evaluate the quality of pliek-u by using an organoleptic test. The ages of the experts were 62, 68 and 70 years; thus, each was well experienced in pliek-u processing. Since the sample numbers 2 and 10 were moldy, only 10 samples were evaluated. The results can be seen in Table 2.

Table 2.  The organoleptic scores of pliek-u collected for the survey.

Sample number	Color	Aroma	Texture	Taste	TOTAL SCORE	AVERAGE SCORE
1	4.0 ± 1	4.0 ± 1	3.7 ± 1.2	4.0 ± 1.7	15.7	4
3	4.0 ± 1	3.0 ± 0	4.0 ± 1	3.7 ± 0.6	14.7	4
4	4.0 ± 1	3.7 ± 0.6	3.3 ± 0.6	4.0 ± 1	15.0	4
5	3.3 ± 0.6	2.7 ± 0.6	3.3 ± 0.6	4.3 ± 0.6	13.7	3
6	3.3 ± 0.6	2.3 ± 1.2	3.3 ± 0.6	3.3 ± 1.2	12.3	3
7	2.7 ± 1.5	2.3 ± 0.6	1.7 ± 0.6	2.3 ± 0.6	9.0	2
8	3.0 ± 1.7	3.3 ± 1.5	4.0 ± 1	4.0 ± 1	14.0	4
9	4.0 ± 1.7	3.0 ± 1	2.7 ± 0.6	3.0 ±1	12.7	3
11	3.0 ± 1.7	2.3 ± 0.6	2.3 ± 0.6	3.7 ± 1.5	11.3	3
12	4.3 ± 1.2	4.3 ± 1.2	4.3 ± 1.2	4.3 ± 1.2	17.3	4
Note: 1 refers to extremely dislike, 2 refers to dislike, 3 refers to neither like nor dislike, 4 refers to like and 5 refers to extremely like; sample numbers 2 and 10 were not evaluated due to their poor quality.

---

 | Show Table

DownLoad: CSV

The highest total score was obtained for sample number 12, at 17.3 (from the district of Aceh Besar), followed by sample number 1, at 15.7 (from the district of Bireuen) and sample number 4, at 15 (from the district of Banda Aceh). However, about five out of 10 samples received average scores of 4 (i.e., "like"). The worst sample was obtained from the district of Bireuen for sample number 7, which had an average organoleptic score of 2 (i.e., "dislike"), as compared to the total score of 9.

From the extensive interviews with the experts, some information about the sensory quality of pliek-u was also documented. Pliek-u has a certain aroma, its color is light to dark brown and its taste can be best described as acidic but not bitter or rancid. All experts agreed on the importance of preserving traditional pliek-u processing in the future. They were consuming pliek-u at a low frequency (i.e., about once in a month), but they hoped for pliek-u to be produced with good and uniform quality by all processors. The taste of pliek-u was found to be very impressive, but they did not have knowledge about the health impact of consuming pliek-u.

3.4.   Color of pliek-u

Figure 2 shows the pictures of the pliek-u samples, and Figure 3 shows the color properties of pliek-u. The incidence of mold can be identified by a white color, and the color of moldy pliek-u appears darker (as can be seen on samples 2 and 10). According to Figure 3, it can be seen that the color properties of pliek-u are varied. The red (R) properties have average of 100 and deviation of 9. The reddest color is sample number 9 with highest R-value of 119. However, the green (G) and blue (B) values are more uniform indicated by lower deviation at 3.5 and 2.1, respectively. Overall, it can be said that the color of these samples was uniform as the deviation of each color properties was lower than its average.

Figure 2.  The samples of pliek-u.

DownLoad: Full-Size Img PowerPoint

Figure 3.  The color properties of pliek-u.

DownLoad: Full-Size Img PowerPoint

3.5.   The volatile compounds of pliek-u

The volatile compounds of the three best samples were analyzed, i.e., sample numbers 1, 4 and 12, as chosen because of their highest total scores for the organoleptic tests (Table 2). It can be seen in Figure 4 that pliek-u contained different volatile compounds. The volatile compounds that were found in all samples were 3-methylbutanal, [ethanol, 2-nitro, propionate (ester)] and (2R, 3R)-butanediol. The volatile compounds found only in sample number 1 were urea (14.59%), azulene (6.72%), cyclic octasulfur (4.75%) and [2-amino-4,5-dimethoxybenzonitrile, N-] (4.72%).

Figure 4.  The volatile compounds of pliek-u: (a) sample number 1, (b) sample number 4 and (c) sample number 12.

DownLoad: Full-Size Img PowerPoint

The volatile compounds identified only in sample number 4 were octanoic acid (8.4%) and [2-thiazolamine, 4-(3,4-dimethoxyphenyl)-5-] (5.9%). The volatile compound found only in sample number 12 was [nitrous acid, ethyl ester] (5.29%). The most volatile compounds were observed in sample number 12. It contained [butanal, 3-methyl] at about 15.496%, methyl formate at 15.013%, silver octanoate at 10.421%, 2-amino-1-pentanol at 6.719% and [2-amino-4,5-dimethoxybenzonitrile, N-] at about 6.307%. Sample numbers 4 and 12 had the same six volatile components, but sample number 12 had fewer volatile compounds compared to sample numbers 1 and 4.

4.   Discussion

The central production of pliek-u in Aceh is Bireuen District. It can be seen that pliek-u processors are rarely found in other parts of Aceh. This warrants attention from universities and the government to conserve this old tradition of coconut endosperm processing. Simple technology can be applied to these small-scale enterprises to improve the situation and attract young people to get involved in preserving the tradition. In addition, the product needs better packaging and labeling in order to be sold, not only in local markets, but also in bigger markets such as supermarkets and online markets. The use of glass material for packaging can also prolong the shelf life of pliek-u ^[21]. So, it is also important to introduce modern packaging methods to processors in Aceh, Indonesia. The short shelf life of the pliek-u in local markets (less than 2 months) is presumed to be associated with high moisture, and the improper storage and packaging of the products.

Considering the organoleptic assessment, five out of the 12 samples received an average score of 4, namely, sample numbers 1, 3, 4, 8 and 12. Using these selected samples, we found the pliek-u samples to have a moisture content of 12–18%, ash content of 5–7%, fat content of 33–46%, protein content of 18–24%, fiber content of 2–3%, carbohydrate content of 13–26% and pH value of 4.0–4.3. In the same manner, based on 3 best samples selected, the expected color of pliek-u is with Red-value ranged from 101 to 113, Green-value ranged from 85 to 94 and Blue-value ranged from 80 to 87.

Based on the PCA results, only five out of seven parameters were used to determine the pliek-u quality. They were the moisture content (${x}_{1})$, pH value (${x}_{2}),$ protein content (${x}_{3})$, ash content (${x}_{4})$ and carbohydrate content (${x}_{5})$. The statistic descriptive of these parameters can be seen in Table A. 1. The excluded parameters were fat content and fiber content. The PCA was conducted by applying the Kaiser-Meyer-Olkin (KMO) value of 0.505 (Table A. 2). The first principal component (${PC}_{1})$ contributed to 34.111% of the variance, with the highest correlation value being attributed to the pH value of about 0.510; the second principal component (${PC}_{2})$ contributed to 24.968% of the variance, with the highest correlation value being attributed to the protein content of 0.537 (The contribution can be seen in Table A. 4 and the correlation value can be seen in Table A. 3). PC₁ and PC₂ can be written as follows (Table A. 5).

Here, it can be seen that the pH value and protein content play the most important roles in determining the quality of pliek-u compared to the other parameters. The importance of the pH value was demonstrated by sample number 4; even though it had a moisture content of 18%, the quality could be maintained without mold. According to the experts, acidity is an important characteristic of pliek-u. The use of a starter culture lowered the pH value from 5.16 to 4.3–4.5 after 3 days of fermentation ^[24]. A similar pH value can be achieved without a starter culture by extending the fermentation process ^[10]. The fermentation step is the core of pliek-u processing because, during this step, the volatile components of pliek-u are also developed. The maximum fermentation was 15 days, according to the age of the lactic acid bacteria, as was determined by Tsukamuto method ^[25]. Further studies are necessary to determine the condition of the fermentation process and its effects on pliek-u quality; improving the technology of the fermentation kettle should be conducted.

Both components (PC₁ and PC₂) had eigenvalues higher than 1, and the cumulative contribution of these components was 59.079%. According to the component plot in rotated space and biplot (Figure 5), the carbohydrate content was located in the same quadrant as the ash content, while the moisture content was located in the same quadrant as the pH value. About five samples were located in quadrant 1, i.e., sample numbers 2, 4, 6, 10 and 11, which had the characteristics of high ash and carbohydrate content. Then, two samples were located in quadrant 2, i.e., sample numbers 7 and 9, which had the same characteristics. Only one sample was located in quadrant 3, i.e., sample number 12, which had no prominent characteristics according to the PCA results, but this sample actually had high fat and protein content (as can be seen in Table 1). Also of importance is the origin of the sample, as sample number 12 was from the Aceh Besar district, which may have caused the different characteristics of the pliek-u. Four samples, i.e., numbers 1, 3, 5 and 8, were located in quadrant 4, indicating the characteristics of a high pH-value and high moisture content. The distribution of samples across all quadrants showed that the quality of the pliek-u highly varied among processors. Therefore, the standardization of pliek-u processing should be formulated.

Figure 5.  The biplot graph of pliek-u quality.

DownLoad: Full-Size Img PowerPoint

Besides the pH value and protein content, the moisture content is commonly important for the quality. This fact was supported by a previous study that indicated that maintaining a moisture content for pliek-u that is below 10% can extend the shelf life up to 2 years ^[2]. Moisture content below 14% is very important for dried products since it can inhibit the growth of microorganisms. It can be the reason why sample numbers 2 and 10, which had moisture contents above 14%, had mold infestation after a week of storage.

The protein content of pliek-u was stable among all processors at 17–24%, with a deviation of about 1.75%. It is interesting that the protein content of pliek-u is higher than the protein content of coconut endosperm. Increasing the protein content was also confirmed in a previous study as the impact the fermentation process ^[26]. Therefore, pliek-u can be used as healthy food due to its high protein content. However, the fat content of pliek-u was highly varied among processors; this could be caused by the method of pressing. High fat content can increase the rancidness during storage of the pliek-u. An appropriate fermentation should increase the crude fat content due to microbial activity producing heat ^[26], but the later separation procedure of oil and pliek-u through the use of pressing technology should reduce the fat content of pliek-u.

The ash content of pliek-u, on average, was 5.41%, with a deviation of 1.37%. It was explained that the fermentation could decrease the ash content, as well as the moisture content ^[26]. A previous study has shown that the ash content of pliek-u was about 4.63% ^[19]. Introducing technology could reduce the contamination during processing and maintain the ash content of pliek-u below 7%.

According to the GC-MS results on volatile compounds, sample numbers 4 and 12 had better components compared to sample number 1 since the urea content in pliek-u sample number 1 was identified. However, sample number 1 also contained beneficial components such as azulene and octasulfur. Azulene is an aromatic hydrocarbon and isomer of naphthalene. Azulene and its derivatives occur naturally as a component of many plants ^[27]. In medicine, the potential use of this compound has been associated with anti-inflammatory, antineoplastic, antidiabetic, antiretroviral with HIV-1 and antimicrobial properties ^[28]. Octasulfur is an inorganic substance that is present in all living tissues. Sulfur is an indispensable food component for human health and disease prevention. Apart from pliek-u, sulfur is also found in abundance in garlic oil, which is often used for health therapy ^[29].

It can be seen that sample number 12 had the best quality since it contained larger amounts of methyl formate; butanal, 3-methyl; octanoid acid; Ag⁺ salt and amino acids. According to the PCA results, sample number 1 was located in quadrant 4, sample number 4 was located in quadrant 1 and sample number 12 was located in quadrant 3. This implies the different characteristics among these samples significantly, which was confirmed later by the differences in volatile compounds. It could be influenced by the unstandardized pliek-u processing. Further studies are recommended to find out the best processing steps in order to produce the beneficial volatile compounds expected from pliek-u.

Finally, pliek-u, as a traditional fermented food, can usually be consumed safely. However, mycotoxins could form during the fermentation of pliek-u ^[11]. A previous study has shown that pliek-u at traditional markets in Aceh had been infected with Aspergillus flavus ^[18]. Although the mycotoxin was not identified in our study, the incidence of urea in sample number 1 was also an example of failure production. Generally, the fermentation of coconut endosperm is predominantly a lactic acid fermentation process ^[4]. To avoid the occurrence of hazardous material in pliek-u, it is important to improve the pliek-u manufacturing process by giving skill and knowledge to the processors and developing simple technologies that can be used in the production. Last but not least, this study has also confirmed that the sensory quality could not stand alone and should be supported by the physicochemical quality to guarantee not only the taste, but also the quality and safety of pliek-u.

5.   Conclusions

Twelve random samples of pliek-u were used to determine the quality of pliek-u produced in Aceh, Indonesia. According to the results of PCA, from the seven parameters evaluated, the most important parameters contributing to pliek-u quality were the pH value (4 to 4.5) and protein content (>19%). However, the three best samples selected by experts were located in different quadrants of the biplot derived from PCA, which revealed that they have different characteristics. The GC-MS results regarding the volatile compounds of these samples confirmed these unique characteristics, which could have been caused by the unstandardized pliek-u processing applied in the community. The step that most significantly affects pliek-u quality is the fermentation step. Further studies are also necessary to determine if a causal relationship exists among these parameters and how certain fermentation conditions would impact the volatile compounds in pliek-u.

Appendix

Table A. 1.  Descriptive statistics of chosen parameters.

	Mean	Std. Deviationa	Analysis Na	Missing N
Moisture	14.8416667	3.07419415	12	0
pH	4.3025000	0.33871750	12	0
Protein	19.8841667	1.75104363	12	0
Ash	5.4116667	1.37156071	12	0
Carbohydrate	22.2375	5.89233574	12	0
Note: a. For each variable, missing values are replaced with the variable mean.

---

 | Show Table

DownLoad: CSV

Table A. 2.  The KMO and Bartlett's tests.

Kaiser-Meyer-Olkin measure of sampling adequacy.		0.505
Bartlett's test of sphericity	Approx. Chi-Square	4.371
	DoF	10
	Sig.	0.929

---

 | Show Table

DownLoad: CSV

Table A. 3.  The anti-image matrices.

		Moisture	pH	Protein	Ash	Carbohydrate
Anti-image covariance	Moisture	0.720	-0.229	-0.038	0.060	-0.328
	pH	-0.229	0.808	0.163	0.211	0.089
	Protein	-0.038	0.163	0.910	0.028	0.186
	Ash	0.060	0.211	0.028	0.906	-0.102
	Carbohydrate	-0.328	0.089	0.186	-0.102	0.747
Anti-image correlation	Moisture	0.508a	-0.300	-0.047	0.074	-0.447
	pH	-0.300	0.510a	0.190	0.247	0.115
	Protein	-0.047	0.190	0.534a	0.031	0.225
	Ash	0.074	0.247	0.031	0.537a	-0.125
	Carbohydrate	-0.447	0.115	0.225	-0.125	0.477a
Note: a. Measures of sampling adequacy (MSA).

---

 | Show Table

DownLoad: CSV

Table A. 4.  The total variance explained by component.

Component	Initial eigenvalues			Extraction sums of squared loadings			Rotation sums of squared loadings
	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %
1	1.706	34.111	34.111	1.706	34.111	34.111	1.617	32.341	32.341
2	1.248	24.968	59.079	1.248	24.968	59.079	1.337	26.739	59.079
3	0.912	18.234	77.313
4	0.684	13.679	90.992
5	0.450	9.008	100.000
Note: Extraction method: principal component analysis.

---

 | Show Table

DownLoad: CSV

Table A. 5.  Rotated Component Matrix^a.

	Component
	1	2
Moisture	0.738	-0.265
pH	0.313	-0.744
Protein	-0.532	0.049
Ash	0.113	0.813
Carbohydrate	0.824	0.221
Note: Extraction method: principal component analysis; Rotation method: quartimax with Kaiser normalization; a. Rotation converged in three iterations.

---

 | Show Table

DownLoad: CSV

Figure A 1.  The scree plot of components.

DownLoad: Full-Size Img PowerPoint

Figure A 2.  The component plot in rotated space.

DownLoad: Full-Size Img PowerPoint

Acknowledgments

We would like to thank the Ministry of Education, Culture, Research and Technology for funding this research through the National Program of Competitive Applied Research (PTKN) for the year 2022 (No. 105/UN11.2.1/PT.01.03/DRPM/2022).

Conflict of interest

We would like to declare that there is no conflict of interest that could affect the publication of this research.