AI-Decoded Multicatalytic Activities Nanozyme Platform: Integrated Identification and Quantification of Multiplexed Pesticide Residues

Analytical array detection holds great promise for multipesticide residue analysis, yet faces key challenges such as cross-channel material interference and error accumulation. Herein, we report a multifunctional "all-in-one" Cu@Zr-MOF nanozyme with intrinsic fluorescence (FL), phosphatase-like (OPH), laccase-like (LAC), and peroxidase-like (POD) activities, which are integrated into a response array for the distinguishing six pesticides ranging from 1.0 to 225.0 ppm via machine learning (ML) technology. This array enables high-throughput discrimination of six pesticides via unsupervised methods. Furtherore, an intelligent stepwise machine-learning approach integrating support vector machine classification and support vector regression achieves both qualitative and quantitative analysis. The entire multichannel analysis is efficient, requiring just 40.0 min to operate (15.0 min for sample-nanozyme preincubation and 25.0 min for signal output). Crucially, the proposed response array enables pesticide detection in six foods with satisfactory recoveries and dynamic monitoring on tomatoes with good accuracy, advancing single-material arrays for food safety monitoring.