Structural Relationship Study of Octanol-Water Partition Coefficient of the Compounds in kesum Essential Oil Using GA-MLR and GA-ANN Methods

Essential Oils are highly concentrated substances the subtle, aromatic and volatile liquids. The use of essential oils is largely widespread in foods, deodorants, pharmaceuticals, drinks, cosmetics, medicine and embalming antiseptics especially with aromatherapy becoming increasingly popular. The lipophilicity of an organic compound can be described by a partition coefficient, logP, which plays a significant role in drug discovery and compound design. A data set of 40 compounds in the essential oil of kesum was randomly divided into 3 groups: training, test and validation sets consisting of 70%, 15% and 15% of data point, respectively. A large number of molecular descriptors were calculated with Dragon software. The Genetic Algorithm - Multiple Linear Regressions (GA-MLR) and genetic algorithm -artificial neural network (GA-ANN) were employed to design the Quantitative Structure-Property Relationship (QSPR) models. The predictive powers of the QSPR model was discussed using Coefficient of determination (R2), Absolute Average Deviation (AAD) and the Mean Squared Error (MSE). The R2 and MSE values of the MLR model were calculated to be 0.734 and 0.194 respectively. The R2 and MSE values for the training set of the ANN model were calculated to be 0.9905 and 2×10-4 respectively. Comparison of the results revealed that the application the GA-ANN method gave better results than GA-MLR method