MOZART, a QSAR Multi-Target Web Based SoftwareBased Tool to Predict Multiple Drug-Enzyme Interactions

Developing models able to predict interactions between drugs and enzymes is a primary goal in computational biology since these models may be used for both predicting new active drugs as well as for the drug repurposing and repositioning. With the compilation of a large dataset of drug-enzyme pairs (66,829), we recognized a unique opportunity to attempt to build a novel multi-target machine learning (MTML) quantitative structure-activity relationship (QSAR) model for probing interactions among different drugs and enzyme targets. For such main aim, we examined here MTML-QSAR models based on the use of topological drugs’ features, along with the artificial neural network (ANN) multi-layer perceptron (MLP). Validation of the final best models found was carried out by internal cross-validation statistics, and other relevant diagnostic statistical parameters. The overall accuracy of the derived model was found to be higher than 96%. Finally, to maximize the diffusion of this model, a public and accessible tool has been developed to allow users performing their own predictions. The tool is able to predict the likely or unlikely interaction between a drug against 23 different enzyme classes targets.