Machine Learning-Based\nNanozyme Sensor Array as an\nElectronic Tongue for the Discrimination of Endogenous Phenolic Compounds\nin Food

The\ndetection of endogenous phenolic compounds (EPs) in food is\nof great significance in elucidating their bioactivity and health\neffects. Here, a novel bifunctional vanillic acid-Cu (VA-Cu) nanozyme\nwith peroxidase-like and laccase-like activities was successfully\nprepared. The peroxidase mimic behavior of VA-Cu nanozyme can catalyze\n3,3′,5,5′-tetramethylbenzidine (TMB) to generate oxidized\nTMB (oxTMB). Owing to the high reducing power of EPs, this process\ncan be inhibited, and the degree of inhibition increases with the\nincrease of reaction time. Additionally, owing to the outstanding\nlaccase mimic behavior of the VA-Cu, it can facilitate the oxidation\nof various EPs, resulting in the formation of colored quinone imines,\nand the degree of catalysis increases with the increase of reaction\ntime. Based on the interesting experimental phenomena mentioned above,\na six-channel nanozyme sensor array (2 enzyme-mimic activities ×\n3 time points = 6 sensing channels) was constructed, successfully\nachieving discriminant analysis of nine EPs. In addition, the combination\nof artificial neural network (ANN) algorithms and sensor arrays has\nsuccessfully achieved accurate identification and prediction of nine\nEPs in black tea, honey, and grape juice. Finally, a portable method\nfor identifying EPs in food has been proposed by combining it with\na smartphone.