Edible Qualities and Flavor Omics of Peanut‐Soybean Dajiang

TL;DR

Researchers developed a peanut-soybean version of Dajiang, a fermented condiment, and found it has improved flavor and taste qualities.

Contribution

The study introduces peanut-soybean Dajiang and identifies unique flavor compounds from its production process.

Findings

Adding peanuts significantly enhanced umami, saltiness, and complex taste in Dajiang.

Eight key flavor compounds were identified, including 2,5-dimethylpyrazine and isoamylphenylacetate.

Changes in amino acid nitrogen, total acid, and other components during fermentation were linked to improved flavor.

Abstract

Dajiang, a fermented bean‐based condiment, is highly popular in Northeast China. In this research, peanut‐soybean Dajiang was successfully developed by incorporating peanuts into the Dajiang production process and refining the fermentation conditions. It was observed that the inclusion of peanuts substantially improved the umami, saltiness, and complex taste profile of Dajiang. Eight key flavor compounds were identified, with 2,5‐dimethylpyrazine and isoamylphenylacetate being unique to peanut‐soybean Dajiang. These findings offer a scientific foundation for enhancing the quality and marketing of peanut‐soybean Dajiang. The study also revealed that alterations in amino acid nitrogen, total acid, fat, reducing sugar, and nitrite levels during fermentation, along with the intensification of umami, salty, and complex flavors, provide a scientific basis for improving the quality of peanut‐soybean Dajiang. The changes in color, aroma, taste, and metabolites througout the fermentation of peanut‐soybean Dajiang were thoroughly analyzed using electronic tongue and electronic nose technologies, in conjunction with the HS‐SPME‐GC–MS non‐targeted metabolomics approach.

Dajiang is a fermented beans‐based condiment that is widely popular in Northeast China. In this study, peanut‐soybean Dajiang was successfully prepared by adding peanut to the Dajiang production process and optimizing the fermentation conditions. It was found that the addition of peanuts significantly enhanced the umami, saltiness and compound taste of Dajiang, and eight key flavor compounds were identified, among which 2,5‐dimethylpyrazine and isoamylphenylacetate were unique to peanut soybean Dajiang. These results provide a scientific basis for the quality improvement and marketing of peanut soybean Dajiang. The study also found that the changes in amino acid nitrogen, total acid, fat, reducing sugar and nitrite content during fermentation, as well as the enhancement of umami, salty and compound taste, provided a scientific basis for the quality improvement of peanut soybean Dajiang. The changes of color, aroma, taste and metabolites during the fermentation of peanut soybean Dajiang were comprehensively analyzed by electronic tongue and electronic nose technology, combined with HS‐SPME‐GC–MS non‐targeted metabolomics method.

Figures

Content of amino acid nitrogen during fermentation. Different letters within the same group indicate significant differences (p < 0.05).

Changes of the response value of the electronic tongue during the fermentation. (a) Ordinary peanut‐soybean Dajiang, (b) high oleic acid peanut soybean Dajiang, (c) soybean Dajiang.

Principal component analysis in fermentation process. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Changes of electronic nose response value during fermentation. (a) Ordinary peanut‐soybean Dajiang, (b) high oleic acid peanut soybean Dajiang, (c) soybean Dajiang.

Principal component diagram of Dajiang. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Loading diagram of three kinds of ma samples.

Sample correlation heat map (a) and Principal Component Analysis (b). P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Upset map of metabolites. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Clustering heat map between Dajiang and metabolites. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Circular map of metabolite species.

KEGG hierarchy (a) and top 20 KEGG metabolic pathways (b) in Dajiang.

OPLS‐DA score plot (a, c, e) and permutation test plot (b, d, f) between peanut‐soybean paste. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Volcanic maps of different metabolites between groups. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Intergroup differential metabolite Venn map (a) and cluster heat map (b). P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Tables

TABLE 1. Changes in flavor‐presenting amino acid content during fermentation.

Fermented time (d)	Sweet amino acids (mg/g)			Bitter amino acids (mg/g)			Savory amino acids (mg/g)			Tasteless amino acids (mg/g)
	P	G	D	P	G	D	P	G	D	P	G	D
0	4.88	3.52	2.68	5.44	3.89	2.76	9.17	6.62	5.66	1.01	0.63	0.34
7	7.59	6.47	6.01	7.60	7.11	5.34	9.91	6.43	7.43	1.30	0.95	0.71
14	8.64	5.67	7.86	7.81	6.1	7.14	9.42	6.61	8.38	1.26	0.84	0.89
21	8.04	8.84	7.65	7.55	7.91	7.6	7.56	8.43	7.06	1.14	1.15	0.92
28	9.51	8.84	11.19	8.71	8.39	11.55	9.17	8.88	11.14	1.26	1.27	1.16
35	8.81	9.37	10.05	8.74	8.39	9.13	7.78	8.17	9.07	1.27	1.23	1.12
42	8.86	9.06	9.30	9.47	8.13	8.9	8.39	7.69	8.86	1.31	1.18	1.11
49	9.65	8.78	9.30	9.05	9.03	9.07	9.18	7.45	8.11	1.34	1.15	1.07
56	12.00	9.80	11.28	10.23	9.52	9.34	10.45	9.02	8.98	1.51	1.35	1.19

TABLE 2. The content of essential amino acids in the fermentation process.

Amino acid (mg/g)	Sample	Fermentation time (d)
		0	7	14	21	28	35	42	49	56
Essential amino acid/E	P	5.65	9.36	10.35	10.03	11.76	11.05	11.38	11.62	14.07
	G	4.35	9.00	7.63	10.78	11.0	11.6	11.11	11.44	12.23
	D	2.99	7.06	9.72	9.78	14.96	12.08	11.68	11.73	13.44
Non‐essential amino acid/N	P	6.94	11.52	12.42	10.95	13.64	12.61	13.85	14.55	17.25
	G	4.44	9.85	8.09	11.81	13.56	12.86	12.17	12.82	14.75
	D	2.38	7.20	10.01	10.21	16.86	13.89	13.51	13.23	14.22
Conditionally essential amino acid	P	1.94	1.60	1.66	1.55	1.71	1.92	2.21	2.03	2.14
	G	1.31	1.33	1.24	1.70	1.52	1.56	1.52	1.92	1.89
	D	1.39	1.42	1.58	1.57	2.41	1.99	1.86	1.85	1.91
Total amino acid/T	P	14.53	22.48	24.43	22.53	27.11	25.58	27.44	28.2	33.46
	G	10.10	20.18	16.96	24.29	26.08	26.02	24.8	26.18	28.87
	D	6.76	15.68	21.31	21.56	34.23	27.96	27.05	26.81	29.57
E/N	P	0.81	0.81	0.83	0.92	0.86	0.88	0.82	0.8	0.82
	G	0.98	0.91	0.94	0.91	0.81	0.9	0.91	0.89	0.83
	D	1.26	0.98	0.97	0.96	0.89	0.87	0.86	0.89	0.95
E/T	P	0.39	0.42	0.42	0.45	0.43	0.43	0.41	0.41	0.42
	G	0.43	0.45	0.45	0.44	0.42	0.45	0.45	0.44	0.42
	D	0.44	0.45	0.46	0.45	0.44	0.43	0.43	0.44	0.45

TABLE 3. Changes in hardness and elasticity during fermentation.

Fermented time (d)	Hardness (N)			Elasticity (mm)
	P	G	D	P	G	D
0	2.63 ± 0.26a	1.46 ± 0.15a	3.14 ± 0.12a	6.96 ± 0.19a	6.26 ± 0.48a	6.46 ± 0.76ab
7	1.7 ± 0.1b	1.26 ± 0.2ab	2.18 ± 0.18b	7.13 ± 0.58a	5.77 ± 0.47ab	8.4 ± 2.77a
14	1.27 ± 0.08cd	1.02 ± 0.23bc	1.02 ± 0.2c	5.23 ± 1.37b	5.26 ± 1.08abc	5.29 ± 0.7bc
21	1.43 ± 0.02c	0.88 ± 0.07c	0.95 ± 0.36c	5.18 ± 0.48b	5.08 ± 1.21abc	5.51 ± 1.75bcd
28	1.11 ± 0.06de	0.47 ± 0.05d	0.5 ± 0.1d	4.55 ± 0.55b	4.71 ± 0.51bc	3.45 ± 0.33cde
35	1.05 ± 0.04e	1.37 ± 0.53ab	0.36 ± 0.04d	4.32 ± 0.44b	4.13 ± 1.4cd	3.15 ± 0.35de
42	0.41 ± 0.11f	0.3 ± 0.03d	0.43 ± 0.08d	2.82 ± 0.86c	2.54 ± 0.57e	2.96 ± 0.3e
49	0.24 ± 0.02fg	0.28 ± 0.05d	0.36 ± 0.04d	2.07 ± 0.19c	2.61 ± 0.28e	3.04 ± 0.46e
56	0.21 ± 0.02g	0.29 ± 0.04d	0.28 ± 0.12d	1.87 ± 0.29c	3.04 ± 0.3de	3.44 ± 0.45cde

TABLE 4. Metabolites of Dajiang.

Category	Serial number	Substance name	CAS	RT (min)	Abundance			Flavor
					P	G	D
Ester	1	Hexadecanoic acid,2‐methyl‐, methyl ester	2490‐53‐1	14.81	5.45 ± 0.18a	5.26 ± 0.12b	5.14 ± 0.08b
	2	Ethyl palmitate	628‐97‐7	14.81	5.49 ± 0.17a	5.23 ± 0.12b	5.08 ± 0.1b	Fruity aroma
	3	isovaleric acid, phenethyl ester	140‐26‐1	19.79	5.11 ± 0.19	—	—
	4	Trimethyl orthoformate Methyl	149‐73‐5	20.57	6.45 ± 0.37a	6.55 ± 0.38a	6.67 ± 0.39a
	5	Ethyl 3‐ (aminomethyl)‐1H‐indole‐2‐carboxylate	1000277‐37‐0	10.79	4.92 ± 0.1	4.85 ± 0.12	—
	6	gamma‐Nonanolactone	104‐61‐0	11.69	5.45 ± 0.03a	5.24 ± 0.02b	5.19 ± 0.03c
	7	Benzylidene diacetate	5062‐30‐6	6.73	—	—	5.14 ± 0.08
	8	3‐Pentanol, 3‐acetate	620‐11‐1	14.8	—	5.1 ± 0.12	—
	9	9, 12‐Octadecadienoic acid (Z,Z)‐, ethyl ester	544‐35‐4	18.84	5.17 ± 0.17a	5.05 ± 0.08a	5 ± 0.14a	Fruity aroma
	10	Methyl dimethoxyacetate	89‐91‐8	20.53	6.47 ± 0.36a	6.56 ± 0.36a	6.96 ± 0.15a
	11	Cyclobutanecarboxylic acid cyclopropyl ester	1000282‐77‐7	1.45	5.29 ± 0a	4.97 ± 0.04c	5.24 ± 0b
	12	Ethyl 9‐cis,11‐trans‐octadecadienoate	1000336‐69‐8	18.85	5.2 ± 0.13a	4.97 ± 0.06b	4.92 ± 0.11b
Alcohols	13	1‐Hexanol	111‐27‐3	5.17	4.92 ± 0.06a	4.85 ± 0.12a	4.56 ± 0.19b	Fruity, floral, and fatty aroma.
	14	Phenethyl alcohol	60‐12‐8	10.25	5.77 ± 0.14a	5.74 ± 0.09a	5.42 ± 0.07a	Rosa rugosa flavor
	15	DL‐1‐Phenethylalcohol	98‐85‐1	10.25	5.81 ± 0.09a	5.72 ± 0.09a	5.44 ± 0.08b
	16	3‐Methyl‐1‐butanol	123‐51‐3	3.99	5.44 ± 0.18	5.21 ± 0.23	—	Unpleasant odor
	17	Isopropanol	67‐63‐0	20.46	5.52 ± 0.04	—	—
	18	1‐Octen‐3‐ol	3391‐86‐4	6.01	6 ± 0.05a	5.66 ± 0.05b	5.97 ± 0.05a	Mushroom, raw soybeans odor
	19	1‐Methylcyclopentanol	1462‐03‐9	7.29	—	4.56 ± 0.11	—
Acid	20	2‐Methylhexanoic acid	4536‐23‐6	7.87	5.43 ± 0.23a	5.24 ± 0.24a	5.41 ± 0a
	21	Acetic acid glacial	64‐19‐7	6.11	5.68 ± 0.34a	5.81 ± 0.3a	5.85 ± 0.32a	Pungent sour odor
	22	Hexanoic acid	142‐62‐1	7.87	5.44 ± 0.28a	5.28 ± 0.18a	5.41 ± 0a	Unpleasant coconut oil odor, spicy taste
Polymeric aldehyde	23	Isovaleraldehyde	590‐86‐3	2.01	5.52 ± 0.13a	5.62 ± 0.15a	5.58 ± 0.12a
	24	Phenylacetaldehyde	122‐78‐1	7.7	5.94 ± 0.06b	5.98 ± 0.07ab	6.05 ± 0.05a	Hyacinth‐like aroma, Diluted sweet fruity fragrance
	25	Benzeneacetaldehyde, a‐ethylidene—	4411‐89‐6	10.5	—	4.91 ± 0.09	5.24 ± 0.1	Honey flavor
	26	Benzaldehyde	100‐52‐7	6.73	5.62 ± 0.1a	5.5 ± 0.08a	5.19 ± 0.08b	Bitter almond flavor, nutty aroma, fruity fragrance
Ketone	27	Benzophenone	119‐61‐9	17.93	4.26 ± 0.08	—	—
	28	3‐Pentanone	96‐22‐0	3.42	—	5.1 ± 0.54	—
	29	3‐Methoxy‐2‐Methyl‐pyran‐4‐one	4780‐14‐7	10.78	4.95 ± 0.08	4.82 ± 0.15	—
Phenol	30	4‐Ethylphenol	123‐07‐9	13.67	5.83 ± 0.01a	5.55 ± 0.04b	5.85 ± 0.04a
	31	2‐Methoxy‐4‐vinylphenol	7786‐61‐0	13.92	5.24 ± 0.04a	5.11 ± 0.1b	4.78 ± 0.05c	Strong spices, clove, and fermented‐like aroma with roasted peanut notes
Hydrocarbon	32	3,5,5‐Trimethyl‐2‐hexene	26456‐76‐8	6.32	4.93 ± 0.12a	4.38 ± 0.33b	5.16 ± 0.13a
	33	3‐Aminophenylacetylene	54060‐30‐9	17.52	5.39 ± 0.05a	5.37 ± 0.06a	5.16 ± 0.08b
	34	Toluene	108‐88‐3	7.69	5.77 ± 0.4a	6 ± 0.06a	6 ± 0.06a
Ether	35	1,3‐Dimethoxybenzene	151‐10‐0	8.61	4.55 ± 0.13c	4.87 ± 0.07b	5.6 ± 0.04a
	36	Dimethyl ether	115‐10‐6	2.06	5.05 ± 0.26b	5.58 ± 0.29a	5.5 ± 0.29a
Heterocyclic compounds	37	2,5‐Dimethyl pyrazine	123‐32‐0	4.99	4.55 ± 0.07	—	—	Strong fried peanuts, and chocolate, creamy smell
	38	2,3,5‐Trimethylpyrazine	14667‐55‐1	5.66	4.73 ± 0.03	4.45 ± 0.07	—	Roasted potatoes, fried peanuts, nuts, earthy aroma, fermented moldy aroma
	39	2‐Pentylfuran	3777‐69‐3	4.18	4.74 ± 0.19a	4.73 ± 0.11a	4.69 ± 0.12a	Ripe soybean flavor, aroma, sweetness
	40	2,3‐Dihydrobenzofuran	496‐16‐2	16.84	5.48 ± 0.04a	5.2 ± 0.05b	4.34 ± 0.01c
	41	3‐Phenylpyridine	1008‐88‐4	14.52	4.73 ± 0.07a	4.91 ± 0.08a	4.82 ± 0.1ab
Other	42	Indole	120‐72‐9	17.51	5.38 ± 0.05a	5.39 ± 0.05a	5.14 ± 0.06b
	43	Ammonium acetate	631‐61‐8	6.38	5.24 ± 0.51a	5.59 ± 0.31a	5.5 ± 0.36a
	44	Quinoxaline,2,3‐dimethyl—	2379‐55‐7	12.26	4.69 ± 0.11	4.57 ± 0	—
	45	Thiourea	62‐56‐6	2.63	5.47 ± 0	—	5.64 ± 0.25
	46	Benzonitrile	100‐47‐0	7.4	4.57 ± 0.04	—	—

TABLE 5. ROAV of Dajiang metabolites.

No.	Flavoring substance	Threshold (mg/kg) a	ROAV
			P	G	D
1	Isovaleraldehyde	0.008	69.95 ± 2.35a	71.33 ± 0.98a	71.37 ± 2.52a
2	2‐Pentylfuran	0.005	100.00	100.00	100.00
3	2,5‐Dimethyl pyrazine	0.02	23.06 ± 0.81	—	—
4	Hexyl alcohol	0.7	0.71 ± 0.03a	0.70 ± 0.01b	0.67 ± 0.03c
5	2,3,5‐Trimethylpyrazine	0.071	6.76 ± 0.29a	6.37 ± 0.11a	—
6	1‐Octen‐3‐ol	0.007	86.80 ± 2.74a	82.11 ± 1.53b	87.33 ± 2.24a
7	Benzaldehyde	0.3	1.90 ± 0.07a	1.86 ± 0.03b	1.77 ± 0.06c
8	Toluene	0.14	4.18 ± 0.28a	4.35 ± 0.10a	4.39 ± 0.12a
9	Phenylacetaldehyde	0.009	66.90 ± 2.81a	67.50 ± 1.39a	68.84 ± 1.54a
10	Hexanoic acid	5	0.11 ± 0.01a	0.11 ± 0.00a	0.11 ± 0.00a
11	4‐Ethylphenol	0.13	4.55 ± 0.18b	4.34 ± 0.11c	4.61 ± 0.10a
12	Indole	0.5	1.09 ± 0.04a	1.09 ± 0.03a	1.05 ± 0.02a
13	Isovaleric acid, phenethyl ester	0.01	51.77 ± 2.50	—	—

Full text

Introduction

1

Dajiang is a traditional condiment originating from Northeast China, prepared through the fermentation of pure soybeans and flour. It is known to reduce cholesterol levels, lower blood pressure, and possess anti‐cancer properties (Xie et al. 2019). The condiment is highly popular in Northeast China and comes in two varieties: those produced through natural fermentation and those using artificial inoculation fermentation. During traditional preparation, there are often no strict operating procedures or specific recipes, which can result in inconsistent quality due to prolonged fermentation. In contrast, industrial production employs artificial inoculation, characterized by controlled temperatures, a shorter fermentation cycle, and a uniform quality (Li et al. 2021). However, this method may affect the unique flavor attributes. In recent years, new soybean products incorporating additional raw materials have been developed to address these issues, aiming to enhance nutritional quality, enrich flavor, and bestow specific functional properties upon Dajiang. For instance, Xiao and colleagues used tartary buckwheat flour and soybeans to create a soybean Dajiang, noting that the addition of tartary buckwheat increased the total acid and total amino acid nitrogen content, thereby improving the overall taste (Longquan Xiao et al. 2022). A study by Chung et al. (2014) demonstrated that brown rice red ginseng miso was effective in improving glucose metabolism and antioxidant defense in mice on high‐fat diets, ‘suggesting that this combination could help lower blood sugar levels and enhance antioxidant activity’.

Peanuts ( Arachis hypogaea Linn.) are widely cultivated in China, with 4684 thousand hectares under cultivation and a production of 18.329 million tons. Peanuts contain various beneficial bioactive components that influence human metabolism and can prevent or alleviate certain diseases (Li Mingrou et al. 2022). Studies have indicated that consuming peanuts is associated with controlling blood pressure, thereby decreasing the occurrence of cardiovascular diseases, heart failure, and myocardial infarction. Additionally, it may reduce the incidence of gallstones, obesity, type 2 diabetes, and cancer (Elissa Haidar et al. 2023). The protein content in peanuts is more similar to that of chickpeas and soybeans than it is to almonds, walnuts, and other oil seeds (Çiftçi and Suna 2022). Furthermore, due to their unique flavor and nutritional content, peanuts are frequently included in food products (Martín et al. 2018). Compounds such as pyrazines and melanoids are produced during the processing of peanuts; these significantly contribute to the flavor of the processed product and are influenced by the method, time, and temperature of processing (López et al. 2024).

Flavor serves as a primary determinant of consumer acceptability and is a crucial indicator of organoleptic quality (Yaxin Gao et al. 2021). The formation of flavor substances is quite complex, influenced by the raw materials used, microbial composition, fermentation cycle, environmental factors, and salinity (Ng'Ong'Ola‐Manani et al. 2013; Zhiluo Que et al. 2023). Yang et al. (2025) utilized a high‐throughput sequencing and headspace solid‐phase microextraction and gas chromatography–mass spectrometry (HS‐SPME‐GC–MS) techniques to explor the bacterial communities and volatile compounds during the fermentation process of Douchiba.

In the present study, peanuts were incorporated into the koji‐making process of Dajiang, prepared using traditional recipes from Northeast China. By employing artificial inoculation with Aspergillus oryzae and Bacillus coagulans , a new variety of Dajiang with superior edible quality was developed. This product addresses the limitation of a singular flavor in traditional Dajiang and reveals that compounds such as 2,3,5‐trimethylpyrazine and isoamyl alcohol are key contributors to enhancing the flavor of peanut‐soybean Dajiang. The findings not only broaden the range of foods that utilize peanuts but also lay the groundwork for improving the nutritional value of Dajiang products, enriching the variety and flavor profiles. Additionally, it meets the diverse needs of consumers and introduces a new method for the efficient and high‐quality production of Dajiang sauce.

Materials and Methods

2

Materials, Reagents, and Equipment

2.1

The peanut varieties Baisha and G965 were provided by Liaoning Zhengye Peanut Production Industry Development Co. Ltd. The soybean raw materials, salt, and flour were purchased from the local farmers' market in Shenyang, Liaoning Province. The chemicals and solvents used for analysis were either analytically pure or high‐performance liquid chromatography (HPLC) grade. Aspergillus oryzae, Bacillus coagulans, and trichloroacetic acid were sourced from Sinopharm Chemical Reagent Co. Ltd. (Beijing, China), whereas formic acid, ammonium formate, and n‐hexane were obtained from Sorabio (Beijing Sorabio Technology Co. Ltd.).

Preparation of the Dajiang and Peanut‐Soybean Dajiang

2.2

Four hundred grams of peanuts and 1600 g of soybeans were rinsed three times with purified water, and then soaked for 12 h. After the red skins were removed from the peanuts, they were steamed and mixed with the soybeans. An additional 300 g of flour was then added and mixed thoroughly. The mixture was cooled to 37°C, inoculated with 0.5% mixed bacteria (Aspergillus oryzae and Bacillus coagulans, 1:1), and then formed into cubes (20 × 10 × 6 cm) which were incubated in a 32°C incubator (Shanghai Yiheng Scientific Instrument Co. Ltd., Shanghai, China) for 28 days and then cut into small pieces (2 × 1 × 1 cm) and placed in a jar. Brine was then added at a volume three times that of the sauce cubes, to achieve a concentration of 12°Bé, followed by stirring 100 times a day and fermentation under natural conditions. The same process was used to make both regular peanut‐soybean Dajiang (P) and high oleic peanut‐soybean Dajiang (G). This method was also used for the preparation of soybean Dajiang (D), utilizing 2000 g of soybeans.

Determination of Amino Acid Nitrogen Content

2.3

The amino acid nitrogen (AAN) contents of the fermented sauces were measured using acid–base titration (Jiang et al. 2007).

Determination of Free Amino Acids

2.4

For sample pretreatment, 141 μL of water and 100 μL of 0.15% Dodecyl Octyl Carbonate were added to 20 mg of sample, followed by 4 μL of an internal standard solutions (Lysine‐d4/Tryptophan‐d5/Glutamine‐d4, 100 μg/mL). The mixtures were sonicated at 40 kHz for 10 min, after which 5 μL of 10 M trichloroacetic acid was added and the samples were frozen and precipitated for 10 min. After centrifugation (High‐speed tabletop refrigerated centrifuge, Shanghai Anting Scientific Instrument Factory, Shanghai, China) (14,000 g, 10 min, 4°C), 25 μL of supernatant was collected and added to 375 μL of water. The mixture was135then vortexed, filtered, and the supematant was collected for further analysis.

For chromatography, an Agilent AdvanceBio MS Spent Media column (2.1 × 50 mm, 2.7 μm) (Agilent Technologies, Santa Clara, CA, USA) was used, with a column temperature of 40°C, and an injection volume of 1 μL. The mobile phase A (was 0.1% formic acid and 10 mM ammonium formate 95% in water), while mobile phase B (was 0.1% formic acid and 10 mM ammonium formate in 95% acetonitrile).

A QTRAP 6500+ system (Sciex, Framingham, MA, USA) in both positive and negative ion modes was used, with 35 psi curtain gas (CUR), medium collision gas (CAD), 50 psi atomization gas (GS1), 50 psi auxiliary heating gas (GS2), spray voltage of 5500 V, and ion source heating temperature of 550°C.

Sample concentrations were calculated by substituing the mass spectral peak area of the sample analyte into the linear equation:

$$[eqn]$$

where C is the concentration measured by LC–MS, V is the volume, DF is the dilution factor, and W is the amount weighed.

Sensory Quality Analysis

2.5

Determination of Texture During Fermentation

2.5.1

A P25 cylindrical probe and texture profile analysis (TPA) mode were used (CT3 10 K texture analyzer, Brookfield, USA) in the hardness and elasticity during Dajiang fermentation. The measurement parameters were a pre‐test speed of 2 mm/s, a test speed of 1 mm/s, a return speed of 1 mm/s, ta rigger force of 0.07 N, and a target value of 10 mm.

Determination of Color During Fermentation

2.5.2

After calibration of the colorimeter (Konica Meta Corporation, Japan), an appropriate amount of evenly ground soybean paste was added to the vessel and spread on the bottom, after which it was placed in the designated position in the colorimeter and covered with a protective cover for color determination. After measurement, the sample was rotated 90° and 180°, respectively, to continue the measurements, and the L*, a*, and b* values of the sample were recorded.

Determination of Taste During Fermentation

2.5.3

Ten grams of evenly ground sample was dissolved in distilled water to 100 mL. The solution was allowed to stand for 2 h at room temperature and then centrifuged at 10000 rpm for 10 min. The supernatant was filtered and placed in a small beaker for electronic tongue testing (SA402B electronic tongue, Insent Inc., Japan), assessing sourness, bitterness, umami, astringency, saltiness, richness, aftertaste‐A (astringent aftertaste), and aftertaste‐B (bitter aftertaste).

Determination of Odor During Fermentation

2.5.4

Ten grams of sample were placed in a 50 mL centrifuge tube, which was then sealed with plastic wrap and capped. The tube was allowed to stand for 15 min to enable even dispersion of the odor, after which the needle of the electronic nose (PEN3 electronic nose, Insent Inc., Japan) was inserted into the tube for measurement. The program settings were self‐cleaning sensor for 90 s and sample analysis for 120 s, with the results obtained after 117–119 s when the system was relatively stable.

Metabolomic Analysis of Flavor Compounds

2.6

Sample Preparation

2.6.1

Approximately 2.0 g of sample was placed in a 20 mL headspace bottle, followed by the addition of 2.0 μL of 100 μg/mL n‐pentadecane‐d32 and immediate sealing of the headspace bottle.

HS‐SPME‐GC–MS Analysis

2.6.2

Preparation of n‐Alkane Standard Solutions

2.6.2.1

Seven hundred and seventy microliters of n‐hexane were placed in a 1.5 mL centrifuge tube, and appropriate amounts of the C10–C25 commercial mixed standard and C26, C27, C28, C29, and C30 n‐alkane standards were added sequentially, followed by vortexing and mixing to obtain a mixed stock solution of C10–C30 n‐alkane (50 μg/mL).

HS‐SPME

2.6.2.2

A 2.0 g sample was placed in a 20 mL headspace bottle and 2.0 μL of internal standard (n‐pentadecane‐d32 100 μg/mL) was added, and the bottle was sealed immediately, followed by equilibration at 80°C for 20 min, aging of the fiber head at 240°C for 5 min, adsorption of the sample for 20 min, and desorption for 2 min.

GC–MS Analysis (Agilent7697A‐8890‐7000D, Agilent Inc., United States)

2.6.3

Quality Control

2.6.3.1

The quality control (QC) sample was a mixture of three types of Dajiang, namely, P, G, and D, all of which received the same treatment. Three QC samples were inserted into the detection process; the repeatability of the QC samples is a reflection of the stability of the instrument during the analytical process, thus ensuring the reliability of the results.

Relative Odor Activity Value (ROAV) Calculations

2.6.4

$$[eqn]$$

where CA is the relative content of the component (%), TA is the sensory threshold of the component (μg/kg), T max is the relative content of the component that contributes the most to the flavor of the sample (%), and C max is the sensory threshold of the component that contributes the most to the flavor of the sample (μg/kg).

Data Analysis

2.7

All experiments were conducted with three replicates. Data were analyzed using SPSS 27 software. Data are presented as mean ± standard deviation, and were compared using ANOVA. Statistical plots and principal component analysis were performed using Origin 2021.

The PEN3 electronic nose software WinMuster was used for principal component analysis and data recording.

Six independent experiments were used for the analysis of the volatile flavor compounds. After the creation of the search library, the matrix files were preprocessed using supplement, normalization, and logarithmic transformation to eliminate or reduce possible errors, and the relative abundance of metabolites was analyzed using SPSS 27 software, with values expressed as mean ± standard deviation.

Results and Analysis

3

Changes in Amino Acid Nitrogen Content During Fermentation

3.1

The AAN content is a key indicator of the quality of Dajiang, which not only reflects the maturity of Dajiang but also its flavor characteristics. As shown in Figure 1, the AAN contents of the three types of Dajiang increased significantly (p < 0.05) with prolongation of the fermentation time, with the highest value of 0.93 g/100 g observed in the G, which is consistent with the results of related studies.

Content of amino acid nitrogen during fermentation. Different letters within the same group indicate significant differences (p < 0.05).

Changes in the Free Amino Acids Content During Fermentation

3.2

Free amino acids are not only important flavor substances but are also precursors of important volatile flavor compounds, and their contents and types directly affect the taste of food (Zhiluo Que et al. 2023). At the same time, the production of amino acids contributes significantly to the maturation of miso (Kim and Lee 2003) The ratios of specific amino acids also represent an important indicator of the nutritional value of the Dajiang.

As can be seen from Table 1, the contents of free amino acids in the three types Dajiang increased over time. On day 56 of fermentation, the contents of alanine, arginine, isoleucine, tyrosine, valine, phenylalanine, glutamic acid, and proline in the two samples with added peanut relative to the D. Arginine is typically found in peanuts, and can regulate serum insulin levels and reduce cholesterol synthesis in the liver (Zhiyuan Tian et al. 2022).

The unique taste of miso results from synergistic interactions between a variety of flavor‐forming amino acids, and thus the relative abundance of specific amino acids has a direct influence on the flavor and taste of miso (Yong Sung Kwon et al. 2019). Amino acids can be classified into four categories according to flavor, namely, sweet, umami, bitter, and tasteless. The contents of these flavor‐associated amino acids in P were ranked as sweet > bitter > umami > tasteless. Both G and P had higher contents of umami amino acids (Table 1).

The contents of essential, non‐essential, and conditionally essential amino acids in the three types of Dajiang increased gradually with fermentation time, with the contents ranked as non‐essential amino acids > essential amino acids > conditionally essential amino acids (Table 2). According to the ideal protein model recommended by Food and Agriculture Organization of the United Nations and the World Health Organization, the E/N ratio of the three samples was higher than 0.6, and the E/T ratio was above 0.4, indicating that all three had high nutritional value.

Sensory Quality Analysis

3.3

Changes in Texture During Fermentation

3.3.1

The texture of Dajiang is an important factor influencing consumer choice. As shown in Table 3, the hardness decreased gradually with the prolongation of fermentation time, and then stabilized—specifically, it dropped from an initial value of 2.63 ± 0.26 N to 0.21 ± 0.02 N before entering the stable stage. This is because proteins, fats, and carbohydrates undergo breakdown into smaller components, making Dajiang progressively less hard. The elasticity also gradually decreased with the extension of the fermentation time—specifically, the elasticity value drops from an initial 6.96 ± 0.19 mm to 1.87 ± 0.29 mm. This is likely due to the evaporation of water during the koji‐making process, resulting in larger voids in the sauce block. This would explain why the elasticity was higher at the beginning of fermentation, and then with the number of rakes increases and the extension of the fermentation time, the sauce block fragments and the elasticity decreased. With the extension of fermentation time, the adhesion gradually decreases—specifically, the adhesion value drops from an initial 1.24 ± 0.08 N to 0.12 ± 0.01 N. This may result from the observation that at the beginning of fermentation, the sauce cubes are larger with incomplete penetration of the brine, resulting in a low moisture content in the sauce cubes. The adhesiveness reflects the amount of energy required to break a semi‐solid substance into an edible state. As a result, P during the early stages of fermentation has greater chewiness (8.67 ± 0.67 N). However, the influence of microbial metabolic activities, the external brine concentration, and the moisture conditions, the hardness of the sauce block decreased, resulting in reduced adhesion. Similarly, over time, the internal structure of the D also undergoes gradual degradation and dispersion, reducing the hardness and also the chewiness.

Changes in Taste During Fermentation

3.3.2

Taste Determination During Fermentation

3.3.2.1

As illustrated in Figure 2, the sensors of the electronic tongue responded to the taste of the Dajiang samples at different fermentation times to different degrees. It was observed that the strongest signals corresponded to saltiness, richness, and umami, whereas the signals for sourness were the weakest. The taste signals for P diminished as the fermentation period lengthened. In summary, the primary flavors of the fermented P were umami, saltiness, and complex tastes, with diminished sourness, bitterness, and astringency.

Changes of the response value of the electronic tongue during the fermentation. (a) Ordinary peanut‐soybean Dajiang, (b) high oleic acid peanut soybean Dajiang, (c) soybean Dajiang.

Principal Component Analysis During Fermentation

3.3.2.2

The principal component analysis (PCA) plots showed that PC1 and PC2 contributed 49.0% and 35.3%, respectively, in the three samples. Figure 3 shows that in terms of PC1, the distance between G and D was relatively large, indicating a significant difference in taste between them, while there was some overlap with D, indicating greater similarity. There is no overlap between P and D, indicating differences in taste. Therefore, use of the electronic tongue can effectively distinguish peanut‐soybean Dajiang from pure soybean Dajiang.

Principal component analysis in fermentation process. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Changes in Odor During Fermentation

3.3.3

Radar Map Analysis During Fermentation

3.3.3.1

The radar map of the electronic nose sensor (Figure 4) shows the overall odor response of the three samples. The 10 sensors in the depicted all indicated varying degrees of response to the samples with differing fermentation times. Notably, the signal response values of the four sensors W5S, W1S, W1W, and W2S were significant and distinct, indicating notable changes in the contents of nitrogen oxides, methyl groups, sulfides, alcohols, aldehydes, and ketones in the samples throughout the fermentation process. The responses of the remaining six sensors exhibited slight overlap during fermentation, suggesting that the volatile substances detected by these sensors remained stable during the process and had essentially similar compositions.

Changes of electronic nose response value during fermentation. (a) Ordinary peanut‐soybean Dajiang, (b) high oleic acid peanut soybean Dajiang, (c) soybean Dajiang.

Principal Component Analysis of Electronic Nose Results in the Three Samples

3.3.3.2

As illustrated in Figure 5, the three samples were significantly separated along PC1, suggesting distinct differences. In terms of PC2, samples G and P were closely clustered, whereas both were far from sample D, indicating a similarity in flavor between the Dajiang with added peanuts and a notable divergence from the Dajiang made exclusively from soybeans. The PCA graph clearly shows no overlap among the three samples, with a high level of discrimination, suggesting that the samples varied in flavor and that the inclusion of peanuts had a significant impact on the flavor profile of Dajiang.

Principal component diagram of Dajiang. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Electronic Nose Load Map Analysis of the Three Samples

3.3.3.3

The analysis of the load map can identify odors that significantly ininfluence in the flavor of Dajiang. In Figure 6, on Day 56, the W5S, W2S, W1S, and W1W sensors contributed the most to PC1, whereas the W1W and W6S sensors made significant contributions to PC2. Regarding the performance characteristics of each sensor, it is evident that PC1 is most sensitive to nitrogen oxides, alcohols, aldehydes, ketones, methyl groups, and inorganic sulfides, while PC2 is sensitive to inorganic sulfides and hydrides.

Loading diagram of three kinds of ma samples.

Metabolomic Analysis of Flavor Compounds

3.4

Comparative Analysis of Samples

3.4.1

The results of the non‐targeted metabolomics samples are presented in Figure 7a, demonstrating strong correlations among samples in the same group. In Figure 7b, the contributions of PCA1 and PCA2 were 23.1% and 38.1%, respectively, with the combined contribution rates totaling 61.2%. The positions of the three samples on the PC1 axis were analyzed, revealing that the components of P and G were similar, with marked differences between them and D. In PC2, the components of D and P were similar, but there was a significant distance between them and G. It is evident that the three samples are distinctly differentiated in the figure, indicating significant differences among them.

Sample correlation heat map (a) and Principal Component Analysis (b). P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Analysis of Metabolites in Dajiang

3.4.2

Metabolites in the Three Samples

3.4.2.1

As illustrated in Figure 8, a total of 42, 40, and 34 metabolites were detected in samples P, G, and D, respectively. Compared to D, the addition of peanuts resulted in a greater number of identified metabolites.

Upset map of metabolites. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Thirty‐one metabolites were common to all three samples, and five metabolites were shared between P and G. One metabolite was common to both P and D, with one metabolite shared between G and D. Five unique flavor substances were found in sample P: 2,5‐dimethylpyrazine, benzonitrile, benzophenone, phenylethyl isovalerate, and isopropanol. There were three unique flavor compounds in sample G: 3‐pentanone, 1‐methylcyclopentanol, and 3‐amyl acetate. D had one unique flavor substance: phenylmethylene diacetate.

To further investigate the relative abundance of metabolites in the individual samples, a cluster heatmap was compiled (Figure 9). The heatmap clearly distinguishes among the three samples: P and G are strongly similar in terms of metabolite abundance and exhibit tight clustering, whereas D is significantly independent of the other two groups, indicating that its metabolite profile is markedly different from the other two. The clustering results further confirm the important role of metabolites in differentiaing pure soybean paste from peanut soybean pastes.

Clustering heat map between Dajiang and metabolites. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

As indicated by Table 4 and Figure 10, esters and alcohols were the two most prevalent compounds, implying that they were the primary contributors to the flavor of Dajiang. The varieties and concentrations of specific compounds varied to differing extents among the three groups of samples. Ester compounds constituted 26.1% of the total volatile flavor compounds; they were the most abundant in the peanut‐soybean Dajiang, imparting fruity, sweet, and ester‐like aromas (Zhiluo Que et al. 2023). For example, ethyl palmitate and ethyl linoleate have fruity aromas (Yulin Lu et al. 2021). Ethyl esters can be used to monitor the effects of miso ripening (Zhao et al. 2011). Alcohols were the second most abundant volatile compounds in soybean paste, contributing refreshing fruity and herbal flavors (Wenwu Ding et al. 2022). Phenylethanol possesses a rose‐like sweetness and is a key aromatic additive in fermented soybean products, including soy sauce, vinegar, and sweet pasta sauces (Sang‐Hee Lee et al. 2021). Within the group of alcohols, 1‐octen‐3‐ol, thurinyl alcohol (1‐phenylethanol) and phenethyl alcohol were found in greater abundance in the D sample. 1‐phenylethanol can serve as a flavoring in food and as an intermediate in pharmaceutical manufacturing (Dong et al. 2017). Studies have claimed that 1‐octen‐3‐ol is responsible for the unpleasant taste of raw soybeans (Zhao et al. 2011). It has also been reported that 1‐octen‐3‐ol is a characteristic product of Aspergillus oryzae spores and is associated with a strong mushroom odor (Peipei Wang et al. 2023). Acids constitute 6.5% of the total volatile flavor compounds; these acidic compounds can be produced through fermentation or ester decomposition by microorganisms such as lactic acid bacteria and yeast in soybean paste (Wei Deng et al. 2022). Aldehydes constituted 8.7% of the total volatile flavor compounds, imparting nutty, green, and fruity aromas (Rong‐Qiang He et al. 2020). They can be produced via lipid oxidation and degradation during fermentation (Ling et al. 2022). Ketones and hydrocarbons each made up 6.5% of the total volatile flavor compounds. Ketones are generally produced through microbial fermentation‐mediated degradation of lipids and amino acids, contributing to the fruity flavor and hazelnut aroma of fermented foods (Weiqi et al. 2023). The flavor profile showed a greater abundance in P and G than in D. Methoxyphenol is a volatile compound that imparts a distinctive smoky taste and phenolic odor to fermented foods. 2‐pentylfuran and 2,3‐dihydrobenzofuran are produced through pyrolysis, and are associated with strong aromas, sweetness, and burnt odors (Sang‐Hee Lee et al. 2021). Pyrazines, which are products of fat oxidation and the Maillard reaction, are characteristic of soybean Dajiang. Their contents may be influenced by cooking conditions, the koji ratio, pasteurization, and bean paste aging (Ling et al. 2022), These compounds have strong aromas of baked potatoes, fried peanuts, and nuts (Weiqi et al. 2023).

Circular map of metabolite species.

Analysis of Flavor Compounds in the Samples

3.4.2.2

The ROAV method is utilized to assess the contributions of individual flavor compounds to the overall flavor profile. Flavor compounds can be categorized based on the magnitude of their ROAV value, with those exhibiting a ROAV of 1 or greater being considered key flavor compounds. Compounds with a ROAV between 0.1 and are deemed to have significant modifying effects on the overall flavor of peanut‐sonbean paste. Table 5 list only metabolites with a ROAV greater than 0.1. The HS‐SPME‐GC–MS analysis identified 11, 9, and 8 metabolites with a ROAV greater than 1 in samples P, G and D, respectively, indicating that these are key contributors to the aroma. Among these, 2‐pentylfuran had the highest ROAV across the three samples. Additionally, two flavor substances exhibited a ROAV between 0.1 and 1.

Table 5 indicates that 8 key flavor compounds with a ROAV greater than 1 were shared among the identified substances. The flavor profile of 2‐pentylfuran is butterscotch and floral; this compound is mainly produced by an enzymatic oxidation reaction catalyzed by lipoxygenase (Weiqi et al. 2023). 1‐octen‐3‐ol, which possesses a mushroom and earthy scent, is formed by enzyme catalysis of unsaturated fatty acids and plays an important role in the flavor production of fermented soy products (Guo et al. 2024). Phenylacetaldehyde imparts a bitter almond flavor, while isovaleraldehyde contributes a malty aroma (Li et al. 2024). Benzoaldehyde is a significant aromatic component in various traditional fermented foods, offering nutty and fruity flavors, such as those found in Doenjang (Menglu Yang et al. 2020). 4‐Ethylphenol is an important phenolic substance in soybean paste and is also thought to have the potential to improve the flavor quality of soy sauce (Shanshan et al. 2022; Zhiluo Que et al. 2023).

KEGG Annotation of Flavor‐Associated Metabolites

3.4.3

The differential metabolites in Dajiang samples were compared and classified using the KEGG database.

As can be seen from Figure 11a, the metabolic pathways in Dajiang can be divided into three types, namely, pathways associated with metabolism, biological systems, and human disease. The most significant metabolism‐associated pathways were related to exogenous biodegradation and amino acid metabolism. As depicted in Figure 11b, the top four metabolic pathways with the highest enrichment were aminobenzoate degradation, associated with two metabolites, benzaldehyde and benzonitrile; phenylalanine metabolism, related to two metabolites, phenylacetaldehyde and phenylethanol; protein digestion and absorption, linked to two compounds, acetic acid and indole; and toluene degradation, associated with two metabolites, benzaldehyde and toluene. During fermentation, benzaldehyde is primarily derived from the metabolism of amino acids by microorganisms, and phenylalanine is broken down by transaminase and decarboxylase to form phenylacetaldehyde, which is then converted into phenylethanol by dehydrogenase, representing the principal metabolic pathway for the production of phenyl ethanol.

KEGG hierarchy (a) and top 20 KEGG metabolic pathways (b) in Dajiang.

Differential Metabolite Analysis

3.4.4

Orthogonal Partial Least Squares Discriminant (OPLS‐DA) Analysis

3.4.4.1

The multivariate supervised orthogonal partial least squares regression analysis (OPLS‐DA) was utilized, with permutation tests employed to ascertain the validity of the outcomes. This approach effectively eliminates external influences to identify flavor differences in peanut‐soybean Dajiang. As illustrated in Figure 12a,c,e, the OPLS‐DA scores revealed that the data from the three groups of samples were positioned on opposite sides of the score chart, suggesting significant differences among the sample groups. R2X and R2Y represent the model's explanatory power for the X and Y matrices, respectively, while Q2 signifies the model's predictive ability. The closer these three values are to 1, the more reliable the results. As depicted in Figure 12b,d,f, the intercept of the Q2 regression line with the vertical axis was less than 0.5 after 200 permutation tests, suggesting no overfitting and confirming the effectiveness of the model validation. Consequently, the results can elucidate the differences in flavor metabolism among the three soybean pastes.

OPLS‐DA score plot (a, c, e) and permutation test plot (b, d, f) between peanut‐soybean paste. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Analysis of Differential Metabolites

3.4.4.2

To understand the differences in metabolites among the three Dajiang samples, a combination of VIP > 1 and p < 0.05 criteria in the OPLS‐DA model was used to identify differential metabolites for each group for further analysis. The results are depicted in the volcano plot (Figure 13).

Volcanic maps of different metabolites between groups. P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

As illustrated in Figure 13, the G vs. D comparison identified a total of 9 differential metabolites, with 7 significantly up‐regulated and 2 significantly down‐regulated. In the G vs. P group, 9 differential metabolites were found, comprising 4 significantly up‐regulated and 5 significantly down‐regulated. The P vs. D group revealed 12 differential metabolites, with 10 being up‐regulated and 2 down‐regulated.

Regarding metabolite categories, the majority of up‐regulated differential metabolites in the G versus D group were alcohols, ketones, esters, and heterocyclic compounds, including isoamyl alcohol, amyl acetate, 3‐pentanone, and 2,3,5‐trimethylpyrazine. Conversely, the down‐regulated compounds were primarily heterocyclic compounds and esters, such as thiourea and phenylmethylenediacetate.

Up‐regulated differential metabolites in the G vs. P group were esters, ketones, aldehydes, and alcohols, including 3‐amyl acetate, 3‐pentanone, 2‐phenyl‐2‐butenal, and 1‐methylcyclopentanol. The down‐regulated differential metabolites included alcohols, esters, other compounds, heterocyclic compounds, and ketones, specifically, isopropanol, phenylethyl isovalerate, benzonitrile, and 2,5‐dimethylpyrazine, among others. Up‐regulated differential metabolites in the P vs. D groups included alcohols, esters, ketones, heterocyclic compounds, and other compounds, such as isopropanol, isoamyl alcohol, phenylethyl isovalerate, 2,3,5‐trimethylpyrazine, and 2,5‐dimethylpyrazine, among others. Twowo differential metabolites were down‐regulated, representing aldehydes and esters, specifically, 2‐phenyl‐2‐butenal and phenylmethylenediacetate.

Figure 14a presents the Wayne diagram, which illustrates the differential metabolites between the peanut and soybean paste groups. In total, 16 differential metabolites were identified across the three soybean paste groups. These metabolites were organized into clustered heatmaps, as depicted in Figure 14b, and comprised esters (4), alcohols (3), ketones (3), aldehydes (1), heterocyclic compounds (2), and other compounds (3). The pure soybean paste exhibited fewer differential metabolites compared to soybean pastes that included added peanuts. Benzophenone, isoamyl alcohol, phenylethyl isovalerate, 2,5‐dimethylpyrazine, and 2,3,5‐trimethylpyrazine are compounds that contribute unique aromas. Among these flavor metabolites, isoamyl alcohol and 2,3,5‐trimethylpyrazine were present in both peanut‐soybean pastes, suggesting that the addition of peanuts imparts sweetness and a peanut‐like aroma, enhancing the overall flavor complexity of the soybean paste.

Intergroup differential metabolite Venn map (a) and cluster heat map (b). P: Ordinary peanut soybean Dajiang, G: High oleic acid peanut soybean Dajiang, D: Soybean Dajiang.

Conclusion

4

In Northeast China, peanuts are incorporated into the koji‐making process of traditional Dajiang preparation. Subsequently, Dajiang483 undergoes fermentation under natural conditions or through artificial inoculation to enhance its quality and flavor. The specific findings are as follows: Throughout the fermentation process, the amino acid nitrogen content of all three samples → the ordinary peanut‐soybean paste, the high oleic acid‐peanut‐soybean paste, and the pure soybean paste → increased progressively. Following fermentation, the levels of amino acids, peptide nitrogen, and eight specific amino acids were elevated in the peanut‐soybean paste, with the concentrations of umami‐related amino acids surpassing those found in the pure soybean paste. Consequently, the inclusion of peanuts improves the taste of Dajiang.

As the fermentation time increased, the predominant flavors became umami, saltiness, and compound tastes, with diminished contributions from sourness, bitterness, and astringency. The taste of the peanut‐soybean Dajiang differed markedly from that of the pure soybean Dajiang, while the regular peanut‐soybean Dajiang and the high oleic acid‐peanut‐soybean Dajiang tasted similar. Differences were also observed in the changes of nitrogen oxides, methyl groups, sulfides, alcohols, aldehydes, and ketones within the peanut‐soybean Dajiang during fermentation. Principal component analysis indicated that there was no overlap between the samples, highlighting significant differences in the odor profiles of the three types of soybean Dajiang and demonstrating that the addition of peanuts had a substantial impact effect on the odor of soybean Dajiang.

Overall, 46 metabolites were detected in the peanut‐soybean Dajiang, with esters and alcohols being the most abundant. There were 8 and 6 additional metabolites in ordinary peanut‐soybean Dajiang and high oleic acid‐peanut‐soybean Dajiang, respectively, compared with pure soybean Dajiang. Among the identified compounds were 8 key flavor compounds with ROAV > 1.2,5‐Dimethylpyrazine and phenylethyl isovalerate were found to be unique key flavor substances in the common peanut‐soybean Dajiang, while 2,3,5‐Trimethylpyrazine was a key flavor substance shared by the common peanut‐soybean Dajiang and the high oleic acid‐peanut‐soybean Dajiang. Nine differential metabolites were identified in the G vs. D group, 9 in the G vs. P group, and 12 in the P vs. D group. Of the 16 differential metabolites selected, esters were the most abundant. Therefore, the addition of peanuts has an effect on the quality of soybean Dajiang.

Incorporating the dietary habits of Northeast China, this research integrates peanuts into the koji‐making process of Dajiang. Through artificial inoculation of bacteria and molds, and fermentation under natural conditions, peanut‐soybean Dajiang is produced. The study investigates the enhancement of quality and flavor in peanut‐soybean Dajiang through this method. This approach not only broadens the processing scope of peanut‐based foods, enhances the nutritional value of Dajiang products, and enriches their variety and flavor, but also meets consumers' diverse demands for taste and nutrition. Thus, it provides a new method for the efficient and high‐quality preparation of Dajiang.

Author Contributions

Yu Miao: conceptualization (equal), funding acquisition (lead), supervision (lead), writing – review and editing (lead). Xu Liwei: conceptualization (equal), writing – original draft (equal). Sun Yu: data curation (lead), writing – original draft (equal). Xie Mengxi: methodology (equal), writing – original draft (equal). Zhang Liangchen: funding acquisition (equal), resources (equal). Yang Hui: validation (equal), writing – review and editing (equal).

Funding

The authors gratefully acknowledge the technical assistance of the food nutrition and quality safety team from LAAS for providing guidance while also allowing the authors to use the necessary instruments. The Project was funded by Liaoning Province key research and development project (2024JH1/10240000402), 2024 Natural Science Foundation of Liaoning Province (2024‐MS‐232) and Discipline Development Project of the Liaoning Academy of Agricultural Sciences (2025XKJS8533).

Conflicts of Interest

The authors declare no conflicts of interest.