Bioinspired Iron\nPorphyrin Covalent Organic Frameworks-Based\nNanozymes Sensor Array: Machine Learning-Assisted Identification and\nDetection of Thiols

Given the crucial role of thiols in maintaining normal\nphysiological\nfunctions, it is essential to establish a high-throughput and sensitive\nanalytical method to identify and quantify various thiols accurately.\nInspired by the iron porphyrin active center of natural horseradish\nperoxidase (HRP), we designed and synthesized two iron porphyrin covalent\norganic frameworks (Fe-COF-H and Fe-COF-OH) with notable peroxidase-like\n(POD) activity, capable of catalyzing 3,3′,5,5′-tetramethylbenzidine\n(TMB) into oxidized TMB with three distinct absorption peaks. Based\non these, a six-channel nanozyme colorimetric sensor array was constructed,\nwhich could map the specific fingerprints of various thiols. Subsequently,\nmachine learning techniques, including supervised learning with linear\ndiscriminant analysis (LDA), decision trees (DT) and artificial neural\nnetworks (ANN), unsupervised learning with hierarchical cluster analysis\n(HCA), and ensemble learning with random forests (RF), were used for\nprecise identification of thiols in complex systems, with a detection\nlimit as low as 50 nM. Significantly, the sensor array demonstrated\nstrong potential for practical applications, including analyzing homocysteine\n(Hcy) in human serum, mercaptoacetic acid (TGA) in depilatory creams,\nand glutathione (GSH) in cell lysates, thereby showing promise for\nuse in disease diagnosis.