Dengue, one of the mosquito-borne viral infections, is a major public health concern across the world. The global incidence of dengue insists the need for a potent antiviral medication for its treatment. Flavonoids are one of the well-known antiviral agents which are being studied with great interest on several viruses including dengue virus. In order to find the structural properties which favor antiviral activity in flavonoids in dengue polymerase, a quantitative structural activity relationship study is performed on a set of 33 flavonoids using multiple linear regression method. Our method showed that the descriptors such as hydrogen bond acceptors, branching index with respect to molecular size and electrotopology of carbon atom explain well the variance in the antiviral activity of flavonoids and the results are statistically significant. The differential interaction pattern observed in docking studies confirms the novelty of the model. The model is applied on a set of flavonoid derivatives, which predicted compounds, 5-hydroxy-3,3',4',6,7,8-hexamethylflavone, 4',5,6,7-tetramethylflavone, and 7″-methylamentoflavone as potent dengue polymerase inhibitors.
Keywords: QSAR; dengue; flavonoids; multiple linear regression; polymerase.