Prediction of solvation free energies of small organic molecules: Additive-constitutive models based on molecular fingerprints and atomic constants

Solvation free energy is an important molecular characteristic useful in drug discovery because it represents the desolvation cost of a ligand binding to a receptor. Most of the recent developments in the estimation of solvation free energy require the use of molecular mechanics and dynamics calculations. Group contribution methods have been rarely used in the past for calculating salvation free energy because automated prediction methods have not been developed in this regard. As an aid to combinatorial library design, we explored rapid and accurate means of computing salvation free energies from the covalent structures of organic molecules and compared the results on a test set with the GB/SA solvation model.. Two independent additive-constitutive QSPR methods have been developed for the computation of solvation free energy. The first is a QSPR model (HLOGS) derived using a technique that uses the counts of distinct/similar fragments and substructures for each molecule as variables in a PLS regression. The second method (ALOGS) uses an extensive atom classification scheme developed earlier for the calculation of Log P. A database of 265 molecules with experimentally determined salvation free energies is used to derive the HLOGS (r = 0.97; rms = 0.58) and ALOGS (r = 0.98; rms = 0.38) models, which were then tested on 27 molecules not present in the training set. A detailed comparison of the HLOGS, ALOGS, GB/SA (with AMBER* and OPLSA* potentials) on the test set showed that the HLOGS and ALOGS models give better results than the GB/SA model. Among the three methods tested, the ALOGS method gives the best result on the test set (r = 0.96; rms = 0.86), though the parametrization for this method is incomplete as many atom types are undetermined due to their absence in the current training set. The HLOGS method appears to handle intramolecular interactions better than the ALOGS method.