Machine learning models identify molecules active against the Ebola virus in vitro

  • Sean Ekins, Joel S. Freundlich, Alex M. Clark, Manu Anantpadma, Robert A. Davey, Peter Madrid
  • F1000Research, October 2015, Faculty of 1000, Ltd.
  • DOI: 10.12688/f1000research.7217.1

Identifying Ebola virus inhibitors using machine learning

What is it about?

This paper assesses machine learning approaches to predict activity against Ebola virus in vitro. 3 compounds were selected using a Bayesian model and all were active in the hundreds of nM range. One of the molecules is an antimalarial called pyronaridine.

Why is it important?

This work is the first example of using Machine learning to suggest compounds to test and is backed up by in vitro testing. Its also important as it shows we can repurpose an EU approved antimalarial as an antiviral. The approach is broadly applicable.

Perspectives

Dr Sean Ekins
Collaborations in Chemistry

In the process of this work we also made the Bayesian models freely accessible so anyone can use them with a mobile app or other open software they develop - for example to score molecules of interest. The next step is to get these compounds into the mouse model.

open access logoRead Publication

http://dx.doi.org/10.12688/f1000research.7217.1

The following have contributed to this page: Dr Sean Ekins and Alex Michael Clark