NMR based metabolomic approach for evaluation of Vietnamese honey

NMR-Based Approaches in the Study of Foods

In this review, the three different NMR-based approaches usually used to study foodstuffs are described, reporting specific examples. The first approach starts with the food of interest that can be investigated using different complementary NMR methodologies to obtain a comprehensive picture of food composition and structure; another approach starts with the specific problem related to a given food (frauds, safety, traceability, geographical and botanical origin, farming methods, food processing, maturation and ageing, etc.) that can be addressed by choosing the most suitable NMR methodology; finally, it is possible to start from a single NMR methodology, developing a broad range of applications to tackle common food-related challenges and different aspects related to foods.

LC-HRMS-Based Non-Targeted Metabolomics for the Assessment of Honey Adulteration with Sugar Syrups: A Preliminary Study

Honey is a natural product that is in great demand and has a relatively high price, thus making it one of the most common targets of economically motivated adulteration. Its adulteration can be obtained by adding cheaper honey or sugar syrups or by overfeeding honeybees with sugar syrups. Adulteration techniques are constantly evolving and advanced techniques and instruments are required for its detection. We used non-targeted metabolomics to underscore potential markers of honey adulteration with sugar syrups. The metabolomic profiles of unadulterated honeys and sugar beet, corn and wheat syrups were obtained using hydrophilic interaction liquid chromatography high-resolution mass spectrometry (LC-HRMS). The potential markers have been selected after data processing. Fortified honey (5%, 10% and 20%), honey obtained from overfeeding, and 58 commercial honeys were analyzed. One potential marker appeared with a specific signal for syrups and not for honey. This targeted analysis showed a linear trend in fortified honeys with a calculated limit of quantification around 5% of fortification.

Identifying type of sugar adulterants in honey: Combined application of NMR spectroscopy and supervised machine learning classification

Nuclear magnetic resonance (NMR) is a powerful analytical tool which can be used for authenticating honey, at chemical constituent levels by enabling identification and quantification of the spectral patterns. However, it is still challenging, as it may be a person-centric analysis or a time-consuming process to analyze many honey samples in a limited time. Hence, automating the NMR spectral analysis of honey with the supervised machine learning models accelerates the analysis process and especially food chemistry researcher or food industry with non-NMR experts would benefit immensely from such advancements. Here, we have successfully demonstrated this technology by considering three major sugar adulterants, i.e., brown rice syrup, corn syrup, and jaggery syrup, in honey at varying concentrations. The necessary supervised machine learning classification analysis is performed by using logistic regression, deep learning-based neural network, and light gradient boosting machines schemes.

•

NMR helps to identify the fingerprints of honey chemical constituents.

•

Combined NMR and ML tools can determine the type of adulteration in honey.

•

Supervised classification schemes, Logistic regression, DNN, and LGBM are utilized.

•

Corn, brown rice, and jaggery adulterations are discriminated in honey.