QSPR modeling of the water solubility of diverse functional aliphatic compounds by optimization of correlation weights of local graph invariants.

The optimization of correlation weights scheme was used to model the water solubility (ln S) of diverse functional aliphatic compounds (n = 193). The optimized descriptor formulated based on the data of a training set (n = 96) generated statistically acceptable relations for the training set (r2 = 0.987), test set (n = 97; r2 = 0.986) and combined set (r2 = 0.987). When the relation of ln S values with the optimized molecular descriptor formulated based on the data of the training set was used for the calculation of ln S values of the training set, r(pred)2 value was found to be satisfactory (0.988), which is indicative of the predictive potential of the scheme. The results indicate the promising potential of the optimization of correlation weights scheme in modeling studies.