Advanced QSRR models of toxicological screening of basic drugs in whole blood by UPLC-TOF–MS

Abstract

A toxicology screen checks a person’s blood or urine or both for the presence of drugs or other toxic substances. The screen can determine the type and amount of drugs or other toxic substances a person may have swallowed, injected, or inhaled. A quantitative structure–retention relationship (QSRR) was developed using the partial least square (PLS), Kernel PLS (KPLS), and Levenberg–Marquardt artificial neural network (L–M ANN) approach for the study of chemometrics. The data which contained retention time (RT) of the 175 toxicological screening of basic drugs in whole blood and tested on authentic samples were obtained by ultra performance liquid chromatography coupled with time-of-flight mass spectrometry. Genetic algorithm (GA) was employed as a factor selection procedure for PLS and KPLS modeling methods. By comparing the results, GA-PLS descriptors are selected for L–M ANN. Finally, a model with a low prediction error and a good correlation coefficient was obtained by L–M ANN. The stability and prediction ability of L–M ANN model were validated using external test set and Y-randomization techniques. The described model does not require experimental parameters and potentially provides useful prediction for RT of new compounds. This is the first research on the QSRR of toxicological screening of basic drugs in whole blood using the chemometrics models.