Virus-Antiviral Prediction Using Machine and Deep Learning Methods

Prediction of new virus-antiviral associations is significant in the treatment of virus diseases. This work proposes a prediction framework for finding new associations among virus diseases and approved antivirals. To achieve this, we have created a new dataset focused on viruses and their corresponding antivirals. Additionally, we propose a prediction framework using random forest, SVM, XGBoost, and a deep learning model. The results are promising and XGBoost outperforms all other methods in the prediction of virus-antiviral associations.