1.What is EBOLApred?

EBOLApred is a machine learning-based webserver for predicting query compounds as active or inactive to inhibit the entry of ebola virus into the host cell

2.How does EBOLApred work?

At the upload page, the interface allows users to paste the smiles format of the compound, identifying it with a molecule ID and simply hitting the submit button for prediction

3.Are there limitations to EBOLApred?

Predictions made on molecules require experimental validation

4.What does confidence score for a prediction stands for?

Confidence score ranges from 0.0 to 1.0. Scores closer to 1.0 means high confidence whilst scores closer to 0.0 is low. So active with 0.87 confidence score means the query is predicted as active but not completely certain as having anti-ebola activity, experimental validation can confirm.

5.What is applicability domain (AD) analysis?

AD for a machine learning model defines an area of chemical space within the training data for which the developed model gives steady and reliable predictions

6.Can prediction results be downloaded?

Results from predictions can be downloaded in a csv file format

7.What machine learning models are used for predictions?

Random forest (RF), Logistic Regression (LR) and Support Vector Machine (SVM) are the available machine learning models for compound predictions.