Entropy multi-target QSAR model for prediction of antiviral drug complex networks

The antiviral QSAR models today have an important limitation. Only they predict the biological activity of drugs against only one viral species. This is determined due the fact that most of the current reported molecular descriptors encode only information about the molecular structure. As a result, predicting the probability with which a drug is active against different viral species only with a single unifying model is a goal of major importance. In this paper we use the Markov Chain theory to calculate new multi-target entropy to fit a QSAR model that predicts by the first time an mt-QSAR model for 500 drugs tested in the literature against 40 viral species. We used Linear Discriminant Analysis (LDA) to classify drugs into two classes as active or non-active against the different tested viral species whose data we processed. The model correctly classifies 1424 out of 1445 non-active compounds (98.55%) and 281 out of 333 active compounds (84.38%). Overall training predictability was 95.89%. Validation of the model was carried out by means of external predicting series, the model classifying, thus, 698 out of 704 non-active compounds and 143 out of 157 active compounds. Overall validation predictability was 97.68%. The present work reports the first attempts to calculate within a unify framework probabilities of antiviral drugs against different virus species based on entropy analysis. We assembled for the first time a drug–virus complex network, for observed possible mechanism of action for the different drugs against viruses.