What is it about?
This is a follow up to our earlier work in which we collated a large set of compounds tested for antitubercular activity in the mouse. We have now added a further 60 molecules from 2014-2015. which were used as a test set for various machine learning models. In addition we have assessed these compounds using mouse liver microsomes and in vitro MTB activity models.
Featured Image
Why is it important?
This work is important because testing compounds in vivo is expensive so any efforts to prioritize them would increase our efficiency. We show that using in vitro models may help in predicting in vivo efficacy. We also provide a novel clustering visualization of all the mouse in vivo TB data to date.
Perspectives
The ability for mouse microsomal T1/2 and in vitro dual event bayesian models to predict in vivo activity data was a novel finding. This suggests that we could expand the applicability domain as we have more of this in vitro data than we do of the in vivo data for model building. The clustering figure provides a single image to visualize all the in vivo data we captured for > 800 compounds
Dr Sean Ekins
Collaborations in Chemistry
Read the Original
This page is a summary of: Machine Learning Model Analysis and Data Visualization with Small Molecules Tested in a Mouse Model ofMycobacterium tuberculosisInfection (2014–2015), Journal of Chemical Information and Modeling, July 2016, American Chemical Society (ACS),
DOI: 10.1021/acs.jcim.6b00004.
You can read the full text:
Contributors
The following have contributed to this page
