Molecular docking aided machine learning for the identification of potential VEGFR inhibitors against renal cell carcinoma.

Abstract

Renal cell carcinoma is a highly vascular tumor associated with vascular endothelial growth factor (VEGF) expression. The Vascular Endothelial Growth Factor -2 (VEGF-2) and its receptor was identified as a potential anti-cancer target, and it plays a crucial role in physiology as well as pathology. Inhibition of angiogenesis via blocking the signaling pathway is considered an attractive target. In the present study, 150 FDA-approved drugs have been screened using the concept of drug repurposing against VEGFR-2 by employing the molecular docking, molecular dynamics, grouping data with Machine Learning algorithms, and density functional theory (DFT) approaches. The identified compounds such as Pazopanib, Atogepant, Drosperinone, Revefenacin and Zanubrutinib shown the binding energy - 7.0 to - 9.5kcal/mol against VEGF receptor in the molecular docking studies and have been observed as stable in the molecular dynamic simulations performed for the period of 500ns. The MM/GBSA analysis shows that the value ranging from -44.816 to -82.582kcal/mol. Harnessing the machine learning approaches revealed that clustering with K = 10 exhibits the relevance through high binding energy and satisfactory logP values, setting them apart from compounds in distinct clusters. Therefore, the identified compounds are found to be potential to inhibit the VEGFR-2 and the present study will be a benchmark to validate the compounds experimentally.