is a web application to predict the metabolic stability of organic compounds, including drug candidates, in silico. Metabolic stability strongly influences a compound’s pharmacokinetic profile, toxicity risks, and overall therapeutic value. The service enables researchers to predict key parameters such as half-life (t1/2) and intrinsic clearance (CLint). While these parameters are traditionally measured in vitro using hepatocytes and liver microsomal fractions, the web tool provides an initial virtual assessment through a convenient interface.
The service is built on classification and regression models trained on data for more than 8,000 chemical compounds collected from ChEMBL (v35) and PubChem. The following algorithms are used: a Naive Bayes classifier with MNA (Multilevel Neighborhoods of Atoms) descriptors, a Self-Consistent Extreme Classifier (SCEC) with QNA (Quantitative Neighborhoods of Atoms) descriptors, and Self-Consistent Regression (SCR) for quantitative models. Quality of prediction includes AUC values above 0.85 for most classification models and R2 up to 0.7 for human CLint regression. The user inputs a compound structure in SMILES format or provides a drug name, after which the service returns two types of predictions:
Qualitative (classification) - the compound is assigned to either the "stable", "unstable" or "moderately stable" class.
Quantitative (regression) - an estimated half-life value for the compound.
A key distinguishing feature of MetaStab-Analyzer is the combination of qualitative and quantitative estimates for three species: human, rat, and mouse.
MetaStab-Analyzer can be used as a rapid in silico tool for the early evaluation of metabolic stability of organic compounds and drug candidates. It helps researchers prioritize molecules with more favorable stability profiles before carrying out labor-intensive and costly experimental studies. The web service is especially useful for medicinal chemists, pharmacologists, and drug discovery teams during hit-to-lead and lead optimization stages. By providing a fast estimate of compound stability, MetaStab-Analyzer supports virtual screening workflows and facilitates more informed decision-making in the selection of promising structures. The MetaStab-Analyzer is part of the Way2Drug tool platform, which also includes PASS Online, MetaTox, and MetaPASS. Together, these tools enable the assessment of the biological activity, metabolism, and toxicity of compounds before synthesis.
Save time and resources - get instant classification and half-life estimates for any compound directly from its SMILES structure, without the need for costly in vitro metabolic assays.
Seamless integration into drug discovery workflows - easily incorporate metabolic stability predictions into your virtual screening pipelines at the hit-to-lead and lead optimization stages.
Part of a comprehensive ADMET analyzer - combine MetaStab-Analyzer results with other free Way2Drug tools (PASS Online, MetaTox, MetaPASS) to build a full ADMET profile of your compounds before synthesis.
Anastasia Rudik et al. (2026)
MetaStab-Analyzer: Classification and Regression Models for Metabolic Stability Prediction.
Molecular Informatics, 45:e70018.
doi: 10.1002/minf.70018
If you need to evaluate a large dataset, or if you need to maintain confidentiality of structural formulas transmitted via unsecured data channels, you can contact us to discuss the licensing opportunities.