Bimetallic ScAg-MOFs as peroxide-mimicking enzymes combined with NIR spectroscopy for rapid prediction of uric acid

In this study, the newly synthesized ScAg-MOFs was used as a like-peroxidase. Based on the positive correlation between uric acid concentration and the (3,3,5,5-tetramethylbenzidine) TMB oxidation inhibition rate, a partial least squares (PLS) model was developed to predict the oxidative inhibition of TMB. The results indicate that the combination of standard normal variable transform and continuous wavelet transform (SNV + CWT) with competitive adaptive reweighted sampling (CARS) yielded a root-mean-square error of cross-validation (RMSECV) of 8.6460%, and a root-mean-square error of prediction (RMSEP) of 7.9731%. The correlation coefficient for the calibration set (R_CV) was 0.9430, for the validation set (R) it was 0.9459, and the prediction deviation ratio (RPD) was 3.0268. This suggests that the model is quite accurate, and it demonstrates the feasibility of this method for rapid uric acid detection.