Prediction of Aqueous Solubility of Drug-Like Compounds by Using an Artificial Neural Network and Least-Squares Support Vector Machine

Abstract

In this work the aqueous solubilities of 145 drug-like compounds were predicted from their theoretical derived molecular descriptors. Descriptors which were selected by stepwise multiple subset selection methods are; 1st-order solvation connectivity index, average span R, overall hydrogen bond basicity, and percent of hydrophilic surface area. These descriptors can encode features of molecules which are effected on dispersion, hydrophobic and steric interactions between solute and solvent molecules. To develop quantitative structure–activity relationship (QSAR) models, the methods of multiple linear regressions, least-squares support vector machine, and artificial neural network (ANN) were used by applying the selected descriptors as their inputs. The obtained statistical parameters of these models revealed that ANN model was superior to other methods. The standard error (SE), average error (AE), and average absolute error (AAE) for ANN model are: SE = 0.714, AE = −0.178, and AAE = 0.546, while these values for internal test set are: SE = 0.830, AE = −0.056, and AAE = 0.630 and for external test set are: SE = 0.762, AE = −0.431, and AAE = 0.626, respectively. Moreover the leave-many-out cross validation test was used to further investigate the prediction power and robustness of model, which lead to RL10O2 = 0.816 and SPRESS = 0.32 for ANN model, which revealed the reliability of this model.

Study on the interaction of anti‐inflammatory drugs with human serum albumin using molecular docking, quantitative structure–activity relationship, and fluorescence spectroscopy

“…In such studies, the observed activity is considered as a dependent variable and molecular descriptors as independent variables. A powerful software for the extraction of a wide variety of molecular descriptors is DRAGON . The DRAGON version 5.5 program calculates 3224 molecular descriptors in 22 classes.…”

“…The obtained model was validated using the leave‐one‐out cross‐validation and Y‐scrambling tests. To assess the robustness of QSAR model, the leave‐one‐out cross‐validation test was applied . In this procedure, first, one row of data (values of descriptors and activity associated to one molecule) was removed, then, the model was regenerated on the remaining dataset, and the activity of the removed molecule was predicted.…”

The experimental and theoretical assessment of biopartitioning micellar liquid chromatography to mimic the drug‐protein binding of some pain‐relief drugs

“…Statistic of Q 2 shows the robustness of the model and should be higher than 0.5, [ 50 ] and SPRESS is the criteria of deviation from experimental data. [ 51,52 ] Some researchers believe that the R 2 and Q 2 statistics are misleading, because prediction errors can be higher for dataset with a wider response range, and vice versa. [ 53 ] Therefore, for further validation of results, mean absolute error (MAE) was proposed along with other error metrics.…”

Machine Learning Techniques and Drug Design

“…Fatemi et al [100] constructed some ANN, SVM and MLR models to predict the aqueous solubility of 145 drug-like compounds. The obtained statistical parameters revealed that the ANN model was superior to other methods, having r 2 equals to 0.816.…”