Prediction of the Partition coefficient of Some anti-HIV drugs by Density Functional Theory

A quantitative structure activity relationship (QSAR) study was performed to develop a model that relates the structures of 26 drug organic compounds to their partition coefficient (log P). Molecular descriptors derived solely from 3D structure were used to represent molecular structures. The compounds are represented by chemical descriptors calculated from their constitutional, geometrical and topological structure, and quantum mechanical wave function. A subset of the calculated descriptors selected using stepwise regression that used in the QSAR model development. Multiple linear regression (MLR) is utilized to construct the linear QSAR model. Stepwise regression was employed to develop a regression equation based on 21 training compounds, and predictive ability was tested on 5 compounds reserved for that purpose. The usefulness of the quantum chemical descriptors, calculated at the level of the DFT theories using 6-31+G** basis set for QSAR study of anti-HIV drugs was examined. The prediction results are in good agreement with the experimental values. A multi-parametric equation containing maximum four descriptors at B3LYP/6-31+G** method with good statistical qualities (R2 train=0.9242, Ftrain=48.768, Q2 LOO=0.8815, R2 adj=0.9052, Q2 LGO=0.8391) was obtained by Multiple Linear Regression using stepwise method

Author(s): Bayat. Z and Emadiyan. M

Abstract | PDF

Share This Article