EBOLApred is a machine learning-based web application for predicting molecules capable of inhibiting the activities of glycoprotein (GP) and the matrix protein (VP40) of the Ebola virus. The developed predictive models were trained on Ebola cell entry inhibitors consisting of 1972 compounds. Random forest (RF) which was the top performing model obtained an F1 score of 0.9 and an area under the curve (AUC) score of 0.95. Users can predict the activity of compounds using random forest (RF), logistic regression (LR) or Support Vector Machine (SVM) models. LR and SVM models also showed plausible performance with overall accuracy of 0.84 and 0.86, respectively.
Adams, J., Agyenkwa-Mawuli, K., Agyapong, O., Wilson, M. D., & Kwofie, S. K. (2022). EBOLApred: A machine learning-based web application for predicting cell entry inhibitors of the ebola virus. Computational Biology and Chemistry, 107766. https://doi.org/10.1016/j.compbiolchem.2022.107766