Artificial Intelligence, Machine Learning, and Big Data for Ebola Virus Drug Discovery.

Area of Science:

• Computational biology
• Drug discovery
• Virology

Background:

• Ebola virus disease (EVD) is a severe, fatal illness requiring new treatments.
• Existing research focuses on identifying biotherapeutic molecules against Ebola virus (EBOV).

Purpose of the Study:

• To review the application of machine learning (ML) techniques for predicting small molecule inhibitors of EBOV.
• To highlight the potential of deep learning models in accelerating anti-EBOV drug discovery.

Main Methods:

• Review of various ML algorithms (Bayesian, SVM, random forest) used for predicting anti-EBOV compounds.
• Discussion on the underutilization and potential of deep neural networks for EBOV drug discovery.
• Summary of essential high-dimensional data sources for ML predictions.

Main Results:

• Established ML algorithms demonstrate credible outcomes in predicting anti-EBOV compounds.
• Deep learning models present an opportunity for developing novel, efficient drug discovery algorithms.
• Comprehensive data is crucial for robust ML-based predictions.

Conclusions:

• ML, particularly deep learning, can significantly enhance EBOV drug discovery efforts.
• AI-driven approaches can improve data-driven decision-making and reduce attrition rates in drug development.
• Leveraging ML can accelerate the search for effective treatments against EVD.