ADME Suite

ADME Suite, developed by ACD/Labs, is a collection of in silico prediction modules designed to calculate pharmacokinetic properties from chemical structure. It supports high-throughput screening of compound libraries, provides insights into pharmacological effects, and helps assess safety for human use. The tool is intended for medicinal, synthetic, and research chemists working in drug discovery and development.

Predictions are based on carefully curated databases of experimental data and cover the full range of absorption, distribution, metabolism, and excretion (ADME) properties. Users can work with single compounds or libraries of tens of thousands of structures, and results include reliability indicators, similar structures from internal databases, and literature references for the underlying experimental data.

Predicted Properties

• Blood-brain barrier penetration (rate of brain penetration logPS, extent logBB, brain/plasma equilibration rate)
• Cytochrome P450 inhibition and substrate specificity for isoforms 3A4, 2D6, 2C9, 2C19, and 1A2
• Distribution, including plasma protein binding (% PPB, logKaHSA) and volume of distribution (Vd) via a physiological Øie-Tozer model
• Maximum recommended daily dose for human clinical trials (oral administration, mg/kg/day)
• Oral bioavailability (%F), including dose-dependence and contributions from solubility, passive absorption, first-pass metabolism, P-gp efflux, and active transport
• Passive absorption across jejunal epithelium and Caco-2 cell monolayers, including transcellular and paracellular route contributions
• P-glycoprotein inhibitor and substrate specificity
• Physicochemical properties including logP, logD, pKa, and aqueous solubility
• Sites of metabolism (regioselectivity)
• Active transport via oligopeptide transporter (PepT1), bile acid transporter (ASBT), monocarboxylic acid transporter MCT1, and amino acid carriers

General Features

• Calculate properties from structures drawn in-app, copied from third-party drawing packages, entered as SMILES strings or InChI codes, or imported from MOL, SK2, SKC, or CDX files, or searched by name in the built-in dictionary
• Automatic detection of tautomeric forms with selection of canonical or major form
• Structure highlighting to indicate substructure and atomic contributions to property values (select modules)
• Batch calculation for groups or libraries of compounds with built-in tools to sort, filter, plot, and rank results
• User-defined label color-coding, numerical filtering, and ascending/descending sorting of results
• Activity history to retrieve previously calculated values
• Results can be reported to PDF or copied to other applications
• Train algorithms with experimental data in select modules: CYP450 Inhibition, Distribution, LogP, LogD, pKa, and P-gp Specificity
• Add custom or in-house prediction algorithms by connecting to an existing web service via XML protocol, or as a DLL (thin client deployments only)

Reliability and Result Assessment

• Each prediction includes a reliability index to gauge result quality
• Display of up to 5 most similar structures from the internal library with experimental values and literature references
• Confidence intervals displayed alongside bar plot results for CYP450 modules
• Alerts for unreliable predictions and for compounds undergoing facilitated diffusion or active efflux across the BBB
• Indication of adverse effects on organs and toxicity in mouse (oral and intravenous) within the Maximum Recommended Daily Dose module

Visualization and Analysis

• Color-mapped structure visualization showing substructure and atomic contributions to property values
• Traffic light indicators for BBB permeability
• Plot of BBB penetration compared against known CNS-active and peripherally-active drugs
• Color-coding of oral bioavailability results to indicate good or poor bioavailability
• Scatter plots, sorting, filtering, and ranking tools for library analysis
• Complete property profile for each molecule accessible in a single interface

Lead Optimization and Structure Modification

• Good/bad indicators for Lipinski's rule-of-5, lead-likeness, cell permeability, plasma protein binding, CNS penetration, metabolic stability, and CYP inhibition
• Interactive optimization tool to modify sets of structures and generate libraries of analogs with substituent modifications targeting an optimal property profile
• Sort, filter, and prioritize hundreds of structural analogs by desired property profile
• Create and use custom fragment libraries
• Built-in retrosynthesis tool to target synthetically accessible fragments
• Ability to alter underlying physicochemical property values (logP, pKa, unbound fraction in plasma) to investigate their effect on BBB permeability, distribution, and absorption

Workflow

1. Draw or import a structure
2. Select the property of interest
3. Review results and make decisions
4. Report results to PDF or copy/paste to another application

ADME Suite is available on the ACD/Labs Percepta platform and supports customization with in-house experimental data to expand the applicability domain of trainable modules. Third-party and in-house models can be incorporated into a single prediction environment. The tool has been adopted by pharmaceutical and academic organizations including Merck KGaA, Loyola University, and Apotex.