Machine learning assisted nanozyme sensor array for accurate identification and discrimination of flavonoids in healthy tea

Abstract

Identifying flavonoids in herbs is of great significance for elucidating their biological activity and pharmacological effects. However, distinguishing and detecting multiple flavonoids simultaneously remains a challenge. Here, an innovative citric acid-Cu (CA-Cu) nanozyme with peroxidase mimic (POD) and laccase mimic (LAC) activities was successfully synthesized. Due to the varying inhibitory effects of flavonoids on CA-Cu dual-enzyme mimicking activities, and the degree of inhibition increasing with prolonged reaction time, a nanozyme sensor array was constructed based on reaction kinetics and applied to the identification of five flavonoids. This technique further streamlines the building of sensing channels. Moreover, by integrating various machine learning algorithms with the sensor arrays, accurate identification and prediction of five flavonoids in multiple herb samples have been successfully achieved. Finally, the sensor array successfully achieved the differentiation and recognition of multiple healthy tea, demonstrating its feasibility in efficiently distinguishing and detecting flavonoids in complex samples.