De-Risking a Large Molecule Through Predictive ADME/Tox and Metabolism Profiling
Case Study 01
De-Risking a Large Molecule Through Predictive ADME/Tox and Metabolism Profiling
A multi-model assessment of off-target effects, transporter liabilities, metabolic stability, and predicted metabolites — delivered as a single decision-ready report.
Challenge
A pharmaceutical company held a large molecule and needed a clear picture of its potential ADME/Tox and off-target effects before committing further resources. Alongside safety and selectivity, they wanted to understand the molecule's metabolic stability and anticipate the metabolites it was likely to form — endpoints that are slow and expensive to explore experimentally up front.
What We Did
-
1
MegaTox machine learning models
Predictions across a broad set of ADME/Tox endpoints.
-
2
MegaTrans machine learning models
Drug-transporter predictions aligned to FDA guidance.
-
3
MegaPredict ChEMBL models
Binding and activity predictions for each human single-protein target drawn from the ChEMBL database.
-
4
MegaPredict SafetyScreen44 models
Virtual analogues of the SafetyScreen44 panel of safety assays.
-
5
MegaPredict BindingDB Search
Similarity-based target identification to surface likely off-target interactions.
-
6
Read-across analysis
Our implementation of the EPA ToxRef database, run in a secure environment.
-
7
Metabolite prediction
A consensus of three metabolite-predictor tools to map likely metabolic fate.
Every prediction was reported with prediction probabilities and applicability domain (AD) scores, so the team knew not just the call but how much to trust it.