ADMET Predictor

ADMET Predictor® is the flagship AI/ML software platform from Simulations Plus, designed for ADMET property prediction, model building, virtual screening, AI-driven drug design, and data analytics. It is built for discovery chemists, DMPK scientists, computational chemists, medicinal chemists, and chemical risk assessment scientists who need to design and evaluate high-quality compounds that are safe, effective, novel, and manufacturable.

The platform quickly and accurately predicts over 175 properties — including aqueous and biorelevant solubility versus pH profiles, logD versus pH curves, pKa (including all microstates), CYP and UGT metabolism outcomes, key toxicity endpoints such as Ames mutagenicity and drug-induced liver injury (DILI) mechanisms, and major systemic pharmacokinetic endpoints via integrated high-throughput GastroPlus® PBPK simulations. Models are trained on premium datasets spanning public and private partner sources, and several have ranked first in independent peer-reviewed comparisons.

Core Prediction Capabilities

• Prediction of over 175 ADMET and physicochemical properties in seconds across large compound sets
• Aqueous and biorelevant solubility versus pH profiles and logD versus pH curves
• pKa prediction including all microstates
• CYP and UGT metabolism outcomes, including metabolite generation and kinetic parameters
• Key toxicity endpoints including Ames mutagenicity and DILI mechanisms
• Integrated high-throughput GastroPlus® PBPK simulations for systemic PK endpoint prediction
• Model applicability domain assessments, confidence estimates, and regression uncertainty for all models
• Customizable visualization tools including distribution plots and 2D/3D scatter plots

Drug Design and Lead Optimization Workflow

• Supports a M.A.P. workflow — Molecular discovery, ADMET prediction, and PK prediction — to integrate ADMET and PK awareness into the earliest design iterations
• Scaffold clustering, R group analysis, and MCDA-based compound prioritization
• Generation of 3D conformations for ligand-based virtual screening against internal or commercial libraries
• Advanced volumetric shape and pharmacophore similarity screening with ADMET- and PK-based triage and ranking of hit selections
• AI-Driven Drug Design (AIDD) module: a generative chemistry and multi-parameter optimization engine that produces novel molecules simultaneously optimized for activity, 3D similarity, ADMET properties, and in vivo PK properties
• Streamlines the design-make-test-analyze (DMTA) cycle and reduces late-stage attrition

Available Modules

• AI-Driven Drug Design (AIDD)
• ADMET Modeler™ — build your own machine learning models or extend existing ones
• High-Throughput Pharmacokinetic (HTPK) Simulation
• MedChem Studio™
• Metabolism
• Physicochemical and Biopharmaceutical (PCB)
• Toxicity
• Transporters
• DILIsym® Inputs Module for mechanistic liver safety risk assessment

Cheminformatics and Data Analysis

• SAR analysis and cheminformatics tools to support lead optimization
• Ability to build custom machine learning models or extend the platform's existing models
• Extensive metabolism prediction capabilities
• Rapid mechanistic liver safety risk assessments via the DILIsym® integration

Enterprise Integration and Automation

• REST API functionality for programmatic access to over 175 property calculations
• Python wrappers and KNIME components for workflow automation
• Data connectors for integration with third-party informatics platforms including Certara D360, Schrödinger LiveDesign®, and BIOVIA Pipeline Pilot
• Supports enterprise-level DMTA cycle workflows across large organizations

Each of the nine modules in ADMET Predictor can be licensed individually, making the platform flexible for teams of varying size and focus. Simulations Plus also offers an Incubator and Biotech Startup Program, on-demand webinars, a tutorial series, and dedicated customer support to help users get the most from the platform.