Flavonoids are a group of naturally occurring phytochemicals abundantly present in fruits, vegetables, and beverages such as wine and tea. In the past two decades, flavonoids have gained enormous interest because of their beneficial health effects such as anti-inflammatory, cardio-protective and anticancer activities. These findings have contributed to the dramatic increase in the consumption and use of dietary supplements containing high concentrations of plant flavonoids. The pharmacological effect of flavonoids is mainly due to their antioxidant activity and their inhibition of certain enzymes. In spite of abundant data, structural requirements and mechanisms underlying these effects have not been fully understood. This review presents the current knowledge about structure-activity relationships (SARs) and quantitative structure-activity relationships (QSARs) of the antioxidant activity of flavonoids. SAR and QSAR can provide useful tools for revealing the nature of flavonoid antioxidant action. They may also help in the design of new and efficient flavonoids, which could be used as potential therapeutic agents.

We report a search for optimum molecular descriptors based on the connectivity index. A suggestion made by several authors that the exponent -0.5 used in the standard formula for computing the connectivity index may not be the optimum for modeling some molecular properties was reexamined. We considered several molecular properties and found that in most cases the optimum value of the exponent is indeed different from -0.5. We suggest that a modified version of the (valence) vertex-connectivity index should be routinely employed in the structure-property modeling instead of the standard version of the index.

No abstract

We present a collection of publicly available intrinsic aqueous solubility data of 829 drug-like compounds. Four different machine learning algorithms (random forests [RF], LightGBM, partial least squares, and least absolute shrinkage and selection operator [LASSO]) coupled with multistage permutation importance for feature selection and Bayesian hyperparameter optimization were used for the prediction of solubility based on chemical structural information. Our results show that LASSO yielded the best predictive ability on an external test set with a root mean square error (RMSE) (test) of 0.70 log points, an R 2 (test) of 0.80, and 105 features. Taking into account the number of descriptors as well, an RF model achieves the best balance between complexity and predictive ability with an RMSE(test) of 0.72 log points, an R 2 (test) of 0.78, and with only 17 features. On a more aggressive test set (principal component analysis [PCA]-based split), better generalization was observed for the RFmodel. We propose a ranking score for choosing the best model, as test set performance is only one of the factors in creating an applicable model. The ranking score is a weighted combination of generalization, number of features, and test performance. Out of the two best learners, a consensus model was built exhibiting the best predictive ability and generalization with RMSE(test) of 0.67 log points and a R 2 (test) of 0.81.

No abstract