Computational modelling of some phenolic diterpenoid compounds as anti-influenza A virus agents

Influenza A virus (IAV) infection is a contagious respiratory disease that causes many deaths due to the advent of drug-resistant strains in recent times. This led to the quest for the in-silico identification of potential hit scaffolds as anti-influenza A agents. In this research, two quantitative structure-activity relationships (QSAR) modelling approaches were performed to relate the molecular descriptors of some phenolic diterpenoids based on their 2D and 3D structural representations with their anti-IAV activities. Subsequently, molecular docking simulation and ADMET evaluations of the compounds were performed to virtually screen and identify the best hits accordingly. The genetic function approximation (GFA) based linear and non-linear regression such as multiple linear regression (MLR) and artificial neural network (ANN) regression models were built in the 2D-QSAR modelling, and the results showed GFA-MLR (R2train = 0.9102, Q2 = 0.8701) and GFA-ANN (R2train=0.9215, Q2=0.9216) models for predicting the anti-IAV activities of the compounds which have passed the global criteria of accepting QSAR models. The 3D-QSAR modelling was carried out based on the comparative molecular field analysis (CoMFA) and comparative similarity indices analysis (CoMSIA), and the results revealed CoMFA_ES (R2train = 0.948, Q2 = 0.590) and CoMSIA_EDH (R2train = 0.980, Q2 = 0.754) models for reliable predictions of anti-IAV activities. The compounds were also virtually screened based on their binding scores through molecular docking with an active site of human hemagglutinin (HA) target which confirms their resilient potency. Furthermore, the drug-likeness and ADMET predictions of the compounds showed the non-violation of Lipinski's rule and good ADMET profiles as part of the rational strategy for future in-silico drug design and discovery. The outcome of this research provides theoretical support to affirm the relevance of totarol as a promising scaffold and set a route for the in-silico design of new diterpenoid inhibitors with improved potency.