Predicting Mouse Liver Microsomal Stability with “Pruned” Machine Learning Models and Public Data

  • Alexander L. Perryman, Thomas P. Stratton, Sean Ekins, Joel S. Freundlich
  • Pharmaceutical Research, September 2015, Springer Science + Business Media
  • DOI: 10.1007/s11095-015-1800-5

Mouse Liver Microsomes models

What is it about?

Mouse liver microsome datasets were curated from the literature and used to build machine learning models. Leave out validation and cross testing was performed.

Why is it important?

Pruning out the moderately unstable / moderately stable compounds from the training set produced models with superior predictive power. Bayesian models displayed the best predictive power for identifying compounds with a half-life ≥1 hour.

Perspectives

Dr Sean Ekins
Collaborations in Chemistry

This work is important because we have cleaned up datasets that can be used by others. These represent the largest public datasets for mouse liver microsomal data. The earlier studies were by big pharma and these data are not publically accessible.

Read Publication

http://dx.doi.org/10.1007/s11095-015-1800-5

The following have contributed to this page: Dr Sean Ekins