Original Articles: 2011 Vol: 3 Issue: 4
Computational approaches to the predication of the octanol-water partition coefficient (LogPo/w)
Abstract
A quantitative structure–activity relationship (QSAR) study was performed to develop models those relate the structures of 41 anti-cancer drugs compounds to their n-octanol–water partition coefficients (log Po/w). Among the different constitutional, topological, geometrical, electrostatic and quantum-chemical descriptors that were considered as inputs to the model. The models were constructed using 33 molecules as training set, and predictive ability tested using 11 compounds. Modeling of log Po/w of these compounds as a function of the theoretically derived descriptors was established by multiple linear regression (MLR). The usefulness of the quantum chemical descriptors, calculated at the level of the HF theories using 6-31G* basis set for QSAR study of anti-cancer drugs was examined. A multi-parametric equation containing maximumeight descriptors at HF/6-31G* method with good statistical qualities (R2 train=0.893, Ftrain=24.93, Q2 LOO=0.816,R2 adj=0.857,Q2 LGO=0.730) was obtained by Multiple Linear Regression using stepwise method.The accuracy of the proposed MLR model was illustrated using the following evaluation techniques: cross-validation, validation through an external test set, and Yrandomisation. The predictive ability of the model was found to be satisfactory and could be used for designing a similar group of compounds.