Discovery of VEGFR2 inhibitors by integrating naïve Bayesian classification, molecular docking and drug screening approaches

Abstract

The high morbidity and mortality of cancer make it one of the leading causes of global death, thus it is an urgent need to develop effective drugs for cancer therapy. Vascular endothelial growth factor receptor-2 (VEGFR2) acts as a central modulator of angiogenesis, and is therefore an important pharmaceutical target for developing anti-angiogenic agents. In this study, ligand-based naïve Bayesian (NB) models and structure-based molecular docking were combined to develop a virtual screening (VS) pipeline for identifying potential VEGFR2 inhibitors from FDA-approved drugs. The best validated naïve Bayesian model (NB-c) gave Matthews correlation coefficients of 0.966 and 0.951 for the test set and external validation set, respectively. 1841 FDA-approved drugs were sequentially screened by the optimal model NB-c and molecular docking module LibDock. By analyzing the results of VS, 9 top ranked drugs with EstPGood value $ 0.6 and LibDock Score $ 120 were chosen for biological validation. VEGFR2 kinase assay results demonstrated that flubendazole, rilpivirine and papaverine showed VEGFR2 inhibitory activities with IC 50 values ranging from 0.47 to 6.29 mM. Binding mode analysis with CDOCKER revealed the action mechanism of the 3 hit drugs binding to VEGFR2. In summary, we not only proposed an integrated VS pipeline for potential VEGFR2 inhibitors screening, but also identified 3 FDA-approved drugs as novel VEGFR2 inhibitors, which could be used to design and develop new antiangiogenic agents.

ADMET profile and virtual screening of plant and microbial natural metabolites as SARS-CoV-2 S1 glycoprotein receptor binding domain and main protease inhibitors

“…Libdock has been widely used as a tool for virtual screening of small molecules against protein or enzyme targets ( Rao et al, 2007 ; Singh et al, 2016 ; Alam and Khan 2018 ). Libdock score which is a cumulative count of all non covalent interactions including the van der Waals has been adopted in many instances for determining the binding affinity (Zhou et al, 2016; Kang et al, 2018 ). Taking the binding affinity against the SARS-CoV-2 main protease into account, putaminoxins B and D, jasmonic acid, jasmonic acid methyl ester were docked against the active site using CDOCKER protocol where the interaction energy of the complex was used as an estimate to select the best binding complex.…”

“…A virtual screening of the selected compounds including a reference antiviral drug penciclovir (PubChem CID: 135398748) ( Razonable 2011 ) was carried out using the DS Libdock, a rigid based program that calculates hotspots for the receptor with placing a grid into the binding site, as well as using polar and apolar probes ( Singh et al, 2016 ; Kang et al, 2018 ). The libdock score determines the binding affinity of the ligands towards a receptor, and is a cumulative count of van der Waals forces, H-bonds, pi interactions and other parameters.…”

“…The docking experiment with libdock was performed setting the ‘Docking Preferences’ user Specified and ‘Max Hits to Save’ to 10. The docked complex with higher libdock score assumed to have higher binding affinity ( Kang et al, 2018 ) and based on the score, docking conformations of selected compounds were optimized using the CDOCKER program which is an implementation of a CHARMm based docking tool ( Gagnon et al, 2014 ). Docking optimization was carried out with a root mean square threshold (RMSD) 0.5 Å and pose cluster radius 0.5 to ensure the docked poses are diverse.…”

Evaluation and Identification of the Neuroprotective Compounds of Xiaoxuming Decoction by Machine Learning: A Novel Mode to Explore the Combination Rules in Traditional Chinese Medicine Prescription

“…Firstly, descriptors in which values appeared in high frequency of more than 50% were eliminated. Secondly, descriptors in which correlation coefficients had an activity of less than 0.1 were excluded [29]. If the absolute value of the correlation coefficients between two descriptors was higher than 0.9, the descriptor possessing a lower correlation coefficient with activity was deleted.…”

“…In this study, the probability values calculated by four models were quoted as new descriptors, and the prediction results were integrated with NB models. Stacked-NB classification (s-NB) models were constructed and validated by DS 2016 [29]. The learning process generated a large set of Boolean features from the input descriptors.…”