A review of conventional and rapid analytical techniques coupled with multivariate analysis for origin traceability of soybean

Figures & data

Figure 1. Demonstrates the pattern of soybean production from 1960 to 2022 (Ritchie and Roser 2021; USDA 2023c).

Figure 2. Outline of analytical techniques discussed in this review.

Table 1. The summary overview of all the chromatographic techniques used to determine soybean’s geographical origin.

Analytical techniques	Geographical origin and sample size	Traceability traits	Data processing techniques	Key results	Ref.
HPLC/PDA	India and Bulgaria	Isoflavones	ANOVA and LSD^2	The isoflavone content of soybeans from either origin did not differ significantly.	(Sakthivelu et al. 2008)
HPLC/DAD	The USA, Russia, Serbia, and China(n = 5 each)	Isoflavones	ANOVA and PCA	Origin had no influence on isoflavone composition. Isoflavone content of different soybean cultivars varied substantially.	(Tepavčević et al. 2010)
HPLC/PDA, GC-FID	The USA (n = 27), China (68), and South Korea (n = 109)	Isoflavones, phenolics, and tocopherol	GLM^3 and LSD	Soybean from these origins could be differentiated based on the given traceability traits.	(Kim et al. 2012)
UPLC-RID, GC, HPLC-UV	The USA (n = 115) and China (n = 218)	Soluble sugars, fatty acids, and total isoflavone	ANOVA, correlation analysis and PCA	Fatty acid, sucrose, and total isoflavone contents were significantly different in soybean from the USA and China.	(Azam et al. 2021)
HPLC-RID, GC-FID	South Korea, The USA, Japan, Taiwan, Nepal, and China	Isoflavone, fatty acids and free sugar	Duncan’s multiple-range test	Soybeans from Nepal could be differentiated from Chinese soybeans based on the isoflavone content.	(Lee and Choung 2011)
GC-FID, HPLC-UV-Vis	The USA (n = 11), China (n = 20), Japan (n = 6) and South Korea (n = 17)	Protein, oil, Fatty acids, and Isoflavones	ANOVA and HCA	Korean soybean had the highest average total isoflavone content.	(Desta et al. 2022)
GC-FID	Four Chinese regions: Beian, Harbin, Jian Sanjiang and Qiqihar	Palmitic acid, heptadecanoic acid, Stearic acid, cis-13-octadecenoic acid and arachidic acid	-NA-	Origin identification rate: 98%. Origin prediction rate: 93%.	(Xi-Chun 2018)
GC-FID	Northern China: Heilongjiang (n = 30), the Inner Mongolia Autonomous Region (n = 20), Jilin (n = 20), Liaoning (n = 21)	Oleic acid and Stearic acid	PCA, HCA and PLS-DA	Origin prediction precision: 90%.	(Sheng et al. 2020)

Table 2. The summary overview of all the mass spectrometry techniques used to determine soybean’s geographical origin.

Analytical techniques	Geographical origin and sample size	Traceability traits	Data processing techniques	Key results	Ref.
IRMS	South Korea (n = 13) and China (n = 10)	δ^15N, δ^13C, δ^18O values	ANOVA and Duncan’s multiple range test	δ^15N values could not be used for origin provenance. No significant difference in the δ^18O values of soybean from South Korea and China.	(Oh et al. 2021)
IRMS	Heilongjiang, Shandong, and Jiangsu provinces of China(n = 18)	δ^15N, δ^13C, values	ANOVA and Duncan’s least significant test	The δ^13C values of soybeans from the three origins varied significantly.	(Wu et al. 2015)
ICP-MS	Vietnam (n = 15), Canada (n = 8), the USA (n = 9) and Brazil (n = 6)	33 trace elements	PCA and HCA	PCA could differentiate soybean from the four origins.	(Nguyen-Quang, Bui-Quang, and Truong-Ngoc 2021)
ICP-MS	Northern China: Heilongjiang (n = 30), the Inner Mongolia Autonomous Region (n = 20), Jilin (n = 20) and Liaoning (n = 21)	12 trace elements	PCA, HCA and PLS-DA	Origin prediction precision: 90%.	(Sheng et al. 2020)
ICP-MS	Chinese Heilongjiang Province: Daqing, Suihua, Heihe and Jiamusi(n = 116)	17 trace elements	OPLS-DA	Origin prediction precision in cross-validation: 93%.	(Cui et al. 2021)
ICP-MS	Argentina provinces: Chaco, Cordoba and San Luis(n = 120)	20 trace elements	SVM-DA	Origin prediction precision in cross-validation: 99%.	(Hidalgo et al. 2020)
HPLC-MS	Southwest China: Sichuan, Chongqing, Guizhou and Yunnan(n = 16)	12 Isoflavones and 8 anthocyanins	Correlation analysis and OPLS-DA	Origin prediction precision: 71%.	(Wu et al. 2017)
HPLC-DAD-ESI/MS	South Korea, USA, Japan, Taiwan, Nepal, and China	Isoflavones	Duncan’s multiple-range test	The isoflavone content in soybean from various sources varied significantly.	(Lee and Choung 2011)
GC-TOF-MS and UPLC-Orbitrap MS	7 Provinces of South Korea	Five metabolites (tryptophan, malonylgenistin, malonyldaidzin, N-acetylornithine, and allysine)	OPLS-DA	Origin prediction rate: 78%.	(Lee et al. 2019)
GC-MS	South Korea (n = 17), China (n = 13) and North America (n = 6)	52 Volatile compounds	PCA, HCA and PLS-DA	There were significant discrepancies between the volatile components of soybean from China and North America.	(Kim et al. 2020)
GC-MS	Argentina, the USA, Indonesia, Australia, Indian, Pakistan, Egypt, China, and Syria(n = 24)	88 Volatile compounds	PCA	-NA-	(Alghamdi et al. 2018)

Table 3. The summary overview of the spectroscopy techniques to determine the geographical origin of soybean.

Analytical techniques	Geographical origin and sample size	Traceability traits	Data processing techniques	Key results	Ref.
^1H NMR	North-eastern, middle, and southern regions of China (n = 9) and South Korea	9 metabolites, 11 metabolites within 3 regions of China, Kernel size	OPLS-DA	Origin prediction rate: 94% Origin classification accuracy: 96%	(Zhou et al. 2021)
Low-field NMR (10 mhz)	China (n = 24) and South Korea (n = 25)	Crude protein and relaxation times	Canonical discriminant analyses	Origin classification accuracy: 96%	(Kim, Rho, and Lee 2009)
Low-field NMR (20 mhz)	South Korea (n = 22) and other countries^4 (n = 34)	Crude fat, Moisture, crude protein, and relaxation times	Long-term discriminant analyses	Origin classification accuracy: 70-92%	(Rho et al. 2003)
FTIR	South Korea (Upper side, Right side, Left side) (n = 8) and China (South-eastern, North-eastern, and Eastern regions) (n = 13)	Three wavenumber regions	PLS-DA and HCA	Predictability: China Vs South Korea: 94%, within China: 89%, and within South Korea: 97%.	(Lee et al. 2018)
NIR	South Korea, China (n = 22), the USA (n = 6), Taiwan (n = 6), Australia (n = 3), Russia (n = 2), Canada (n = 2), Brazil (n = 1), and Ukraine (n = 1)	-NA-	PCA and MPLS	Origin prediction rate: 98%	(Ahn and Kim 2012)
FT-NIR	South Korea (n = 234) and other countries^1 (n = 237)	Eight wavenumber regions	PLS	Origin prediction rate: 96%	(Lee et al. 2022)
NIR and ED-XRF	South Korea (n = 37) and other countries^1 (n = 75)	Al, P, S, Cl, Ca, Fe, Ni, Cu, Zn, Br and Rb	PCA, MPLS, and canonical DA	Correct classification rate: 99%	(Kim et al. 2022)
ED-XRF	South Korea (n = 100) and imported (USA, China, Vietnam, and Thailand; n = 97)	Five trace elements (P, Ni, Br, Zn and Mn)	Canonical DA	Accuracy in discriminating south Korean soybean from other countries: 100%. Accuracy in discriminating soybean within south Korea: 93%	(Lee et al. 2020)
ED-XRF	Northern China: Henan, Inner Mongolia, Xinjiang, Heilongjiang, and Liaoning(n = 296)	Nine trace elements (Mg, K, Ca, Mn, Fe, Ni, Cu, Zn, and Rb)	PCA and MLP	Origin classification accuracy: 96%	(Lai et al. 2020)
ED-XRF with 3 D polarization optics	Japan (n = 23), the USA (n = 8), Canada (n = 7), and China (n = 7)	Eight trace elements	PCA and LDA	Origin prediction precision: 91%	(Otaka, Hokura, and Nakai 2014)
Terahertz spectroscopy	Argentina (n = 25), the USA (n = 9), and China (n = 41)	-NA-	iPLS, ELM, GA-SVM, ABC-SVM	Origin prediction rate: 95%	(Wei, Zhu, et al. 2020)

Ritchie, H., and M. Roser. 2021. Forests and deforestation. Our World in Data. Published Online at OurWorldInData.org. Retrieved from: https://ourworldindata.org/forests-and-deforestation.

 Google Scholar

Accessed October 23, 2022.

 Google Scholar

USDA. 2023c. World Agricultural Production. Foreign Agricultural Service. WAP 1-23. USA: United States Department of Agriculture.

 Google Scholar

Sakthivelu, G., M. K. Akitha Devi, P. Giridhar, T. Rajasekaran, G. A. Ravishankar, M. T. Nikolova, G. B. Angelov, R. M. Todorova, and G. P. Kosturkova. 2008. Isoflavone composition, phenol content, and antioxidant activity of soybean seeds from India and Bulgaria. Journal of Agricultural and Food Chemistry 56 (6):2090–5. doi: 10.1021/jf072939a.

 PubMed Web of Science ®Google Scholar

Tepavčević, V., M. Atanacković, J. Miladinović, D. Malenčić, J. Popović, and J. Cvejić. 2010. Isoflavone composition, total polyphenolic content, and antioxidant activity in soybeans of different origin. Journal of Medicinal Food 13 (3):657–64. doi: 10.1089/jmf.2009.0050.

 PubMed Web of Science ®Google Scholar

Kim, E.-H., H.-M. Ro, S.-L. Kim, H.-S. Kim, and I.-M. Chung. 2012. Analysis of isoflavone, phenolic, soyasapogenol, and tocopherol compounds in soybean [Glycine Max (L.) Merrill] germplasms of different seed weights and origins. Journal of Agricultural and Food Chemistry 60 (23):6045–55. doi: 10.1021/jf300463f.

 PubMed Web of Science ®Google Scholar

Azam, M., S. Zhang, J. Qi, A. M. Abdelghany, A. S. Shaibu, S. Ghosh, Y. Feng, Y. Huai, B. S. Gebregziabher, J. Li, et al. 2021. Profiling and associations of seed nutritional characteristics in Chinese and USA soybean cultivars. Journal of Food Composition and Analysis 98:103803. doi: 10.1016/j.jfca.2021.103803.

 Web of Science ®Google Scholar

Lee, J. H., and M.-G. Choung. 2011. Comparison of nutritional components in soybean varieties with different geographical origins. Journal of the Korean Society for Applied Biological Chemistry 54 (2):254–63. doi: 10.3839/jksabc.2011.040.

 Google Scholar

Desta, K. T., O. S. Hur, S. Lee, H. Yoon, M.-J. Shin, J. Yi, Y. Lee, N. Y. Ro, X. Wang, and Y.-M. Choi. 2022. Origin and seed coat color differently affect the concentrations of metabolites and antioxidant activities in soybean (Glycine Max (L.) Merrill) seeds. Food chemistry 381:132249. doi: 10.1016/j.foodchem.2022.132249.

 PubMed Web of Science ®Google Scholar

Xi-Chun, L. U. 2018. Study on origin identification traceability based on contents of soybean fatty acid content components [Language of text: Chinese]. Food Research and Development 39 (16):55–9.

 Google Scholar

Sheng, C. D., J. H. Yu, L. H. Qing, W. Z. Hui, and M. Xue Fei. 2020. Geographical specificity of fatty acid and multi-element fingerprints of soybean in Northern China. Quality Assurance and Safety of Crops & Foods 12 (3):126–39. doi: 10.15586/qas.v12i3.767.

 Web of Science ®Google Scholar

Oh, H.-J., G-a Ko, M.-Y. Yang, and Y.-J. Kim. 2021. 안정동위원소비를 활용한 고춧가루와 콩의 원산지 판별 분석 [Stable isotope ratio analysis for origin authentication of red pepper powders and soybeans]. Journal of the Korean Society of Food Science and Nutrition 50 (6):570–6. doi: 10.3746/jkfn.2021.50.6.570.

 Google Scholar

Wu, Y., D. Luo, H. Dong, J. Wan, H. Luo, Y. Xian, X. Guo, F. Qin, W. Han, L. Wang, et al. 2015. Geographical origin of cereal grains based on element analyser-stable isotope ratio mass spectrometry (EA-SIRMS). Food chemistry 174:553–7. doi: 10.1016/j.foodchem.2014.11.096.

 PubMed Web of Science ®Google Scholar

Nguyen-Quang, T., M. Bui-Quang, and M. Truong-Ngoc. 2021. Rapid identification of geographical origin of commercial soybean marketed in Vietnam by ICP-MS. Journal of Analytical Methods in Chemistry 2021:e5583860. doi: 10.1155/2021/5583860.

 PubMed Web of Science ®Google Scholar

Cui, D., Y. Liu, H. Yu, Z. Wang, and X. Mao. 2021. Geographical traceability of soybean based on elemental fingerprinting and multivariate analysis. Journal of Consumer Protection and Food Safety 16 (4):323–31. doi: 10.1007/s00003-021-01340-2.

 Web of Science ®Google Scholar

Hidalgo, M. J., D. C. Fechner, D. Ballabio, E. J. Marchevsky, and R. G. Pellerano. 2020. Traceability of soybeans produced in argentina based on their trace element profiles. Journal of Chemometrics 34 (12):e3252. doi: 10.1002/cem.3252.

 Web of Science ®Google Scholar

Wu, H-j., J-c Deng, C-q Yang, J. Zhang, Q. Zhang, X-c Wang, F. Yang, W-y Yang, and J. Liu. 2017. Metabolite profiling of isoflavones and anthocyanins in black soybean [Glycine Max (L.) Merr.] Seeds by HPLC-MS and geographical differentiation analysis in Southwest China. Analytical Methods 9 (5):792–802. doi: 10.1039/C6AY02970A.

 Web of Science ®Google Scholar

Lee, E. M., S. J. Park, J.-E. Lee, B. M. Lee, B. K. Shin, D. J. Kang, H.-K. Choi, Y.-S. Kim, and D. Y. Lee. 2019. Highly geographical specificity of metabolomic traits among Korean domestic soybeans (Glycine Max). Food research International (Ottawa, Ont.) 120:12–8. doi: 10.1016/j.foodres.2019.02.021.

 PubMed Web of Science ®Google Scholar

Kim, M.-J., J. Lim, S. W. Kwon, G. Kim, M. S. Kim, B.-K. Cho, I. Baek, S. H. Lee, Y. Seo, and C. Mo. 2020. Geographical origin discrimination of white rice based on image pixel size using hyperspectral fluorescence imaging analysis. Applied Sciences 10 (17):5794. doi: 10.3390/app10175794.

 Google Scholar

Alghamdi, S. S., M. A. Khan, E. H. El-Harty, M. H. Ammar, M. Farooq, and H. M. Migdadi. 2018. Comparative phytochemical profiling of different soybean (Glycine Max (L.) Merr.) genotypes using GC–MS. Saudi journal of Biological Sciences 25 (1):15–21. doi: 10.1016/j.sjbs.2017.10.014.

 PubMed Web of Science ®Google Scholar

Zhou, Y., S.-Y. Kim, J.-S. Lee, B.-K. Shin, J.-A. Seo, Y.-S. Kim, D.-Y. Lee, and H.-K. Choi. 2021. Discrimination of the geographical origin of soybeans using NMR-based metabolomics. Foods 10 (2):435. doi: 10.3390/foods10020435.

 PubMed Web of Science ®Google Scholar

Kim, M.-H., J.-H. Rho, and C.-H. Lee. 2009. NMR relaxation 기법을 이용한 한국산과 중국산 대두의 원산지 판별 [The geographical discrimination of Korean and Chinese Soybeans (Glycine max (L.) merrill) using NMR relaxation methods] [text in Korean]. Korean Journal of Food Science and Technology 41 (3):292–5.

 Google Scholar

Rho, J.-H., S.-M. Lee, Y.-B. Kim, and T.-S. Lee. 2003. 20 MHz pulsed NMR을 이용한 국내산과 수입산 콩의 판별 [Discriminating domestic soybeans from imported soybeans by 20 MHz pulsed NMR] [text in Korean]. Korean Journal of Food Science and Technology 35 (4):653–9.

 Google Scholar

Lee, B.-J., Y. Zhou, J. S. Lee, B. K. Shin, J.-A. Seo, D. Lee, Y.-S. Kim, and H.-K. Choi. 2018. Discrimination and prediction of the origin of Chinese and Korean soybeans using fourier transform infrared spectrometry (FT-IR) with multivariate statistical analysis. PloS One 13 (4):e0196315. doi: 10.1371/journal.pone.0196315.

 PubMed Web of Science ®Google Scholar

Ahn, H.-G., and Y.-H. Kim. 2012. 근적외선분광광도계(NIRS)를 이용한 국내산 콩과 수입콩의 판별분석 [Discrimination of Korean domestic and foreign soybeans using near infrared reflectance spectroscopy]. Korean Journal of Crop Science 57 (3):296–300. doi: 10.7740/kjcs.2012.57.3.296.

 Google Scholar

Lee, J. H., J. M. An, H. J. Kim, H. C. Shin, S. H. Hur, and S. H. Lee. 2022. Rapid discrimination of the country origin of soybeans based on FT-NIR spectroscopy and data expansion. Food Analytical Methods 15 (12):3322–33. doi: 10.1007/s12161-022-02375-3.

 Web of Science ®Google Scholar

Kim, N., M. Jang, J. Jo, J. Park, A. Kim, and I. Hwang. 2022. Application of energy dispersive X-ray fluorescence spectrometry and near-infrared reflectance spectroscopy combined with multivariate statistical analysis for discriminating the geographical origin of soybeans. Food Control. 140:109140. doi: 10.1016/j.foodcont.2022.109140.

 Web of Science ®Google Scholar

Lee, J.-H., D.-J. Kang, E.-H. Jang, S.-H. Hur, B.-K. Shin, G.-T. Han, and S.-H. Lee. 2020. ED-XRF (Energy dispersive X-ray fluorescence spectrometer)를 이용한 콩 원산지 판별[Discrimination of geographical origin for soybeans using ED-XRF] [text in Korean]. Korean Journal of Food Science and Technology 52 (2):125–9. doi: 10.9721/KJFST.2020.52.2.125.

 Google Scholar

Lai, H., J. Xi, J. Sun, W. He, Z. Wang, C. Zheng, and X. Mao. 2020. Multi-elemental analysis by energy dispersion X-ray fluorescence spectrometry and its application on the traceability of soybean origin. Atomic Spectroscopy 41 (1):20–8. doi: 10.46770/AS.2020.01.003.

 Web of Science ®Google Scholar

Otaka, A., A. Hokura, and I. Nakai. 2014. Determination of trace elements in soybean by X-ray fluorescence analysis and its application to identification of their production areas. Food chemistry 147:318–26. doi: 10.1016/j.foodchem.2013.09.142.

 PubMed Web of Science ®Google Scholar

Wei, X., W. Zheng, S. Zhu, S. Zhou, W. Wu, and Z. Xie. 2020. Application of terahertz spectrum and interval partial least squares method in the identification of genetically modified soybeans. Spectrochimica acta. Part A, Molecular and Biomolecular Spectroscopy 238:118453. doi: 10.1016/j.saa.2020.118453.

 PubMed Web of Science ®Google Scholar