Pharmaceutical Research

, Volume 33, Issue 2, pp 433–449

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

  • Alexander L. Perryman
  • Thomas P. Stratton
  • Sean Ekins
  • Joel S. Freundlich
Research Paper

DOI: 10.1007/s11095-015-1800-5

Cite this article as:
Perryman, A.L., Stratton, T.P., Ekins, S. et al. Pharm Res (2016) 33: 433. doi:10.1007/s11095-015-1800-5



Mouse efficacy studies are a critical hurdle to advance translational research of potential therapeutic compounds for many diseases. Although mouse liver microsomal (MLM) stability studies are not a perfect surrogate for in vivo studies of metabolic clearance, they are the initial model system used to assess metabolic stability. Consequently, we explored the development of machine learning models that can enhance the probability of identifying compounds possessing MLM stability.


Published assays on MLM half-life values were identified in PubChem, reformatted, and curated to create a training set with 894 unique small molecules. These data were used to construct machine learning models assessed with internal cross-validation, external tests with a published set of antitubercular compounds, and independent validation with an additional diverse set of 571 compounds (PubChem data on percent metabolism).


“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 h.


Our results suggest the pruning strategy may be of general benefit to improve test set enrichment and provide machine learning models with enhanced predictive value for the MLM stability of small organic molecules. This study represents the most exhaustive study to date of using machine learning approaches with MLM data from public sources.

Graphical Abstract

Key Words

Bayesian model machine learning metabolic stability mouse liver microsomal stability translational research 



Absorption metabolism, distribution, excretion and toxicity


Collaborative Drug Discovery


Molecular function class fingerprints of maximum diameter 6


Human liver microsomal stability


High Throughput Screens


Mycobacterium tuberculosis


positive predictive value


Quantitative Structure-Activity Relationships


Receiver-operator characteristic


Structure Activity Relationship


Support Vector Machine

Supplementary material

11095_2015_1800_MOESM1_ESM.docx (2.3 mb)
Supplementry Material 1(DOCX 2331 kb)
11095_2015_1800_MOESM2_ESM.sdf (2.1 mb)
Supplementry Material 2(SDF 2101 kb)
11095_2015_1800_MOESM3_ESM.sdf (2.2 mb)
Supplementry Material 3(SDF 2211 kb)
11095_2015_1800_MOESM4_ESM.sdf (59 kb)
Supplementry Material 4(SDF 59 kb)
11095_2015_1800_MOESM5_ESM.sdf (2.8 mb)
Supplementry Material 5(SDF 2877 kb)
11095_2015_1800_MOESM6_ESM.sdf (3.3 mb)
Supplementry Material 6(SDF 3370 kb)

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  • Alexander L. Perryman
    • 1
  • Thomas P. Stratton
    • 2
  • Sean Ekins
    • 3
    • 4
  • Joel S. Freundlich
    • 1
    • 2
  1. 1.Division of Infectious Disease, Department of Medicine, and the Ruy V. Lourenço Center for the Study of Emerging and Re-emerging PathogensRutgers University–New Jersey Medical SchoolNewarkUSA
  2. 2.Department of Pharmacology & PhysiologyRutgers University-New Jersey Medical SchoolNewarkUSA
  3. 3.Collaborations in ChemistryFuquay-VarinaUSA
  4. 4.Collaborative Drug DiscoveryBurlingameUSA

Personalised recommendations