Ebola Virus Bayesian Machine Learning Models Enable New in Vitro Leads

  • Manu Anantpadma, Thomas Lane, Kimberley M. Zorn, Mary A. Lingerfelt, Alex M. Clark, Joel S. Freundlich, Robert A. Davey, Peter B. Madrid, Sean Ekins
  • ACS Omega, January 2019, American Chemical Society (ACS)
  • DOI: 10.1021/acsomega.8b02948

Machine learning for Ebola drug discovery

What is it about?

Our previously published Ebola models from 2015 were rebuilt with Assay Central, tested with external datasets and used to select additional compounds for testing. This extends our earlier work in that this time we tested many more compounds and also looked at different .

Why is it important?

We used a couple of papers by others to provide external test sets and demonstrate that the machine learning models would have identified many of these compounds as active. Several further in vitro active compounds were identified e.g. Arterolane and Lucanthone.

Perspectives

Dr Sean Ekins
Collaborations in Chemistry

The compounds identified extend our earlier work from 2015 and demonstrate how other antimalarials (e.g. arterolane) appear to have activity against this virus. We also evaluated the various compounds for PAINS.

Read Publication

http://dx.doi.org/10.1021/acsomega.8b02948

The following have contributed to this page: Dr Sean Ekins