Abstract:In the current work, an empirical solvation model based on accessible surface areas was reported, which can be used to predict the solvation free energies for both organic and biological molecules very fast. This solvation model is based on atom-weighted solvent accessible surface area (SAWSA). The parameterization procedure for different kinds of atoms was performed as follows: first, the atoms in a molecule were defined to different atom types based on SMARTS language; then the solvent accessible surface area for each atom (or charged group) was calculated; finally, 4 genetic algorithm (GA) was applied to optimize the solvation parameters for different atom types in order to reproduce the experimental solvation free energies. The derived model possessed promising predictive ability as indicated by the good statistical significance and good prediction on the external test set. Using the solvation model based on all 377 neutral molecules, we have achieved an average unsigned error of 0:51 kcal/mol and standard deviation of 0.46 kcal/mol, which was better than the model proposed by Wang et al. The solvation model developed in the current work was applied to predict the solvation free energies of small organic molecules and proteins. For the,51 small organic molecules, the SAWSA model could give consistent results with the AM1/SM2.1 model in addition to several molecules with large conjugate systems. Moreover, the predictions from SAWSA were much better than those from SM5.0R, a solvation model based on geometry-dependent atom surface tensions. For the IS proteins randomly selected from the Brookhaven PDB database, the solvation free energies predicted by the SAWSA model showed high linear correlation (r = 0.99) than those predicted by PBSA, which were much better than those given by the Ooi model and the Vial model. Finally, we have successfully applied this model to predict the relative binding free energies for four binding modes of EGFR/quinazoline. The most favorable binding mode identified by MM-PBSA could also be correctly recognized by MM-SAWSA. The relative solvation free energies calculated by SAWSA show obvious correlation with those calculated by PBSA. The SAWSA should have potential applications in QSAR, molecular docking, protein folding, free energy calculations, and so forth.

Assessing Solvate Prediction Approaches: A Case of Spironolactone

Insights into Intermolecular Interactions of Spironolactone Solvates

Solubility prediction for cholesterol and desmosterol in mixed solvents via NRTL-SAC and COSMO-RS

Development of prothioconazole solvates by prediction and experiment to enhance solubility and stability

Computational screening for prediction of co-crystals: method comparison and experimental validation

Prediction of cosolvent effect on solvation free energies and analysis of intermolecular forces of tizanidine hydrochloride in ethyl acetate + methanol / ethanol mixtures

Predicting drug solubility in organic solvents mixtures: A machine-learning approach supported by high-throughput experimentation

Crystallization solvent design based on a new quantitative prediction model of crystal morphology

Using COSMO-RS to Predict Hansen Solubility Parameters

Structural Insights into the Highly Solvating System of Axitinib via Binary and Ternary Solvates

Computational prediction of drug solubility in lipid based formulation excipients

Research progress of solubility prediction through quantitative structure-property relationship

Enhancing Predictions of Drug Solubility through Multidimensional Structural Characterization Exploitation

Holistic Prediction of the Pka in Diverse Solvents Based on a Machine‐Learning Approach

Modeling the solubility of pharmaceuticals in pure solvents and solvent mixtures for drug process design

A Robust Crystal Structure Prediction Method to Support Small Molecule Drug Development with Large Scale Validation and Prospective Studies

Qspr Model Analysis On The Solubility Of Organic Compounds In Ionic Liquids

Electrostatic Potential as Solvent Descriptor to Enable Rational Electrolyte Design for Lithium Batteries

Prediction of the hydrophobicity of platinum(IV) complexes based on molecular surface properties

Empirical Aqueous Solvation Models Based on Accessible Surface Areas with Implicit Electrostatics

Scaling Solute-Solvent Distances to Improve Solubility and Ion Paring Predictions in Rigid Ion Models