Related Experiment Video
Updated: Apr 2, 2026
Human Liver Microphysiological System for Assessing Drug-Induced Liver Toxicity In Vitro
Published on: January 31, 2022
Predicting Mouse Liver Microsomal Stability with "Pruned" Machine Learning Models and Public Data
Alexander L Perryman1, Thomas P Stratton2, Sean Ekins3,4
1Division of Infectious Disease, Department of Medicine, and the Ruy V. Lourenço Center for the Study of Emerging and Re-emerging Pathogens, Rutgers University-New Jersey Medical School, Newark, New Jersey, 07103, USA.
Purpose:
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.
Methods:
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).
Results:
"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.
Conclusions:
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.
Insights
Machine learning models can predict mouse liver microsomal stability, aiding drug discovery. A pruning strategy improved model accuracy for identifying stable drug compounds.
Area of Science:
- Pharmacology and Toxicology
- Computational Chemistry
- Drug Discovery
Background:
- Mouse efficacy studies are crucial for translational research but are preceded by metabolic stability assessments.
- Mouse liver microsomal (MLM) stability studies are an initial, though imperfect, model for predicting metabolic clearance.
- Identifying compounds with good MLM stability is essential for advancing potential therapeutics.
Purpose of the Study:
- To develop machine learning (ML) models for predicting MLM stability.
- To enhance the identification of compounds with favorable metabolic stability profiles.
- To improve the efficiency of early-stage drug discovery pipelines.
Main Methods:
- Compiled a training dataset of 894 unique small molecules with published MLM half-life values from PubChem.
- Constructed ML models, including Bayesian approaches, using the curated dataset.
- Assessed model performance through internal cross-validation, external testing with antitubercular compounds, and independent validation with 571 diverse compounds.
Main Results:
- A "pruning" strategy, removing moderately stable compounds, significantly improved model predictive power.
- Bayesian ML models demonstrated the highest accuracy in identifying compounds with a half-life of ≥1 hour.
- The models showed enhanced predictive value for MLM stability.
Conclusions:
- The pruning strategy offers a generalizable method to improve test set enrichment for MLM stability.
- Machine learning models, particularly Bayesian, provide enhanced predictive value for the MLM stability of small organic molecules.
- This study represents a comprehensive application of ML to publicly available MLM data for drug development.
