Quantitative structure-property relationship prediction of gas-to-chloroform partition coefficient using artificial neural network

A quantitative structure-property relationship (QSPR) study based on an artificial neural network (ANN) was carried out for the prediction of the gas-to-chloroform partition coefficients of a set of 338 compounds of a very different chemical nature. The genetic algorithm-partial least squares (GA-PLS) method was used as a variable selection tool. A PLS method was used to select the best descriptors and the selected descriptors were used as input neurons in neural network model. These descriptors are Gravitation index for all bonded pairs of atoms (G2), Final heat of formation (ΔHf), Total hybridization components of the molecular dipole (µh), DPSA-3 Difference in CPSAs (DPSA-3) and Structural Information content (order 1) (1SIC). The results obtained showed the ability of developed artificial neural networks to predict of gas-to-chloroform partition coefficients of various compounds. Also this demonstrates the advantages of ANN.