Prediction of solubility of some statin drugs in supercritical carbon dioxide using classification and regression tree analysis and adaptive neuro-fuzzy inference systems

Abstract

A quantitative structure-solubility relationship was developed to predict the solubility of some statin drugs in supercritical carbon dioxide (SC-CO2). The solubility of lovastatin, simvastatin, atorvastatin, rosuvastatin, and flovastatin in SC-CO2 at 225 different states of temperature and pressure were predicted. Classification and regression tree (CART) was successfully used as a descriptor selection method. Three descriptors (pressure, temperature, and molecular weight) were selected and used as inputs for adaptive neuro-fuzzy inference system (ANFIS). The root mean square errors for the calibration, prediction, and validation sets were 0.09, 0.14, and 0.11, respectively. In comparison with other methods, CART-ANFIS is a powerful model for prediction of solubilities of these statins in SC-CO2.