QSAR Prediction of Aqueous Solubility's of Some Pharmaceutical Compounds by Chemometrics Methods

A quantitative structure–activity relationships (QSAR) study is suggested for the prediction of solubility of pharmaceuticalcompounds in aqueous solution by using chemometrics methods. Ab initio theory was used to calculate some quantum chemicaldescriptors including electrostatic potentials and local charges at each atom, HOMO and LUMO energies, etc. Also, Dragonsoftware was used to calculate some descriptors such as WIHM and GETAWAY. QSAR studies are mathematical quantification ofrelations between structure and activity or property. These are extensively used in pharmaceutical and agricultural chemistry forscreening potential compounds for specific biological activity. Computable molecular descriptors are preferred to experimentalproperties in QSAR analyses because require molecular structure as the only input and can be in expensively calculated for achemical in less than a millisecond. By multivariate calibration methods such as partial least squares (PLS) regression and leastsquares support vector analysis (LS-SVM), it is possible to obtain a model adjusted to the concentration values of the mixtures usedin the calibration range. Orthogonal signal/descriptor correction (OSC/ODC) is a preprocessing technique used for removing theinformation unrelated to the target variables based on constrained principal component analysis. OSC is a suitable preprocessingmethod for PLS calibration of mixtures without loss of prediction capacity using cited descriptors. The root mean square error ofprediction (RMSEP) was also quite acceptable for OSC-PLS (0.0095) and LS-SVM (0.0023)