What is it about?
This work collates a large library of over 600 molecules tested vs TB Topo I and uses them to build various machine learning models. These modes were compared vs docking for identifying new inhibitors.
Featured Image
Why is it important?
While we can generate statistically meaningful machine learning models after 5 fold cross validation etc, these models were not successful in identifying new compounds for in vitro testing. Docking in a homology model seemed to do much better. A more recently derived apo structure for TB top I was used to dock some of out in vitro actives to rationalize them. We also compared the homology model vs the apo structure.
Perspectives
There may be benefits to having different approaches to trying to identify additional inhibitors of TB Topo I. We have to date predominantly focused on FDA approved drugs but both machine learning and docking could be used to filter other libraries of molecules before testing.
Dr Sean Ekins
Collaborations in Chemistry
Read the Original
This page is a summary of: Machine learning and docking models for Mycobacterium tuberculosis topoisomerase I, Tuberculosis, March 2017, Elsevier,
DOI: 10.1016/j.tube.2017.01.005.
You can read the full text:
Contributors
The following have contributed to this page
