Prediction of Surface Tension of Propane Derivatives Using QSPR Approach and Artificial Neural Networks

Abstract

: A Quantitative Structure-Activity Relation study (QSPR) was conducted to evaluate the surface tension of a series of 30 propane derivatives. The surface tension (TS) of these derivatives was correlated with a single calculated descriptor, namely (Mor13v).In the present work, multiple linear regression (MLR) and artificial neural network (ANN) techniques were used for QSPR studies of the tension surface  of 30 propane derivatives.The results of MLR model were compared with those of the ANN model. the comparison showed that the R² = 96.58%, s = 1.4178 of ANN were higher and lower, respectively, which illustrates that an ANN presents an excellent alternative for developing a QSPR model for the surface tension (ST) values of propane derivatives compared to MLR.