Comparison of various methods for validity evaluation of QSAR models

BACKGROUND

Quantitative structure-activity relationship (QSAR) modeling is one of the most important computational tools employed in drug discovery and development. The external validation of QSAR models is the main point to check the reliability of developed models for the prediction activity of not yet synthesized compounds. It was performed by different criteria in the literature.

METHODS

In this study, 44 reported QSAR models for biologically active compounds reported in scientific papers were collected. Various statistical parameters of external validation of a QSAR model were calculated, and the results were discussed.

RESULTS

The findings revealed that employing the coefficient of determination (r²) alone could not indicate the validity of a QSAR model. The established criteria for external validation have some advantages and disadvantages which should be considered in QSAR studies.

CONCLUSION

This study showed that these methods alone are not only enough to indicate the validity/invalidity of a QSAR model.

In silico modeling studies of N9-substituted harmine derivatives as potential anticancer agents: combination of ligand-based and structure-based approaches

A computational study was carried out to develop quantitative-structure activity relationship (QSAR), pharmacophore, molecular docking and molecular dynamics simulations of a series of N9-substituted harmine derivatives in order to investigate the structural factors involved in the cytotoxic activity and thus design new active derivatives. A valid 3 D-QSAR (R²= 0.89, q²=0.67, R²pred = 0.72) and 2 D-QSAR (R²= 0.81, q²=0.69, R²pred = 0.76) models were obtained correlating the cytotoxic activity with hydrophobic and hydrogen bond acceptor (HBA) features for 3 D-QSAR and SlogP and a_acc descriptors for 2 D-QSAR. Analysis of the selected descriptors for both models highlighted that lipophilicity and hydrogen bonding acceptor atoms remain the crucial properties and those on which cytotoxic activity depends. Also, these findings are in agreement with the characteristics of the generated pharmacophore. Furthermore, molecular docking revealed that the binding energy (-9.74 kcal/mol) and inhibition constant (0.071 µmol) correlate with the activity of the most active compound that forms hydrophobic interactions and two hydrogen bonds with the the dual specificity tyrosine phosphorylation regulated kinase 1 A (DYRK1A). The molecular dynamics simulations revealed that the protein-ligand equilibrium is stable after 100000 fs of trajectories. Based on these results, we designed new N⁹-substituted harmine derivatives with improved properties: predicted cytotoxic activities, estimated binding energies, estimated inhibition constants and interaction modes with amino acid residues of DYRK1A, compared to the best compound in the studied dataset. Additionally, these newly designed inhibitors showed promising results in the preliminary in silico Absorption, Distribution, Metabolism, Excretion and Toxicity (ADMET) evaluations.Communicated by Ramaswamy H. Sarma.