MathWorks

SimBiology is a MathWorks product that provides an app and programmatic tools for modeling, simulating, and analyzing dynamic biological systems. It is designed for applications in quantitative systems pharmacology (QSP), physiologically-based pharmacokinetic (PBPK), and pharmacokinetic/pharmacodynamic (PK/PD) modeling. Users can build models interactively through the SimBiology block diagram editor or programmatically using the MATLAB language. Models can be created from scratch, imported as SBML-formatted files, or based on built-in model examples.

SimBiology supports a range of analytical techniques for ODE-based models of varying complexity. Users can run simulations to assess target feasibility, predict drug efficacy and safety, and identify optimal dosing schedules. Parameter estimation is supported through nonlinear regression and nonlinear mixed-effects (NLME) methods, and non-compartmental analysis (NCA) can be performed on both experimental and simulation data.

Model Building

• Models can be constructed visually in the SimBiology Model Builder, in a manner similar to drawing on paper, or imported from existing SBML files.
• Dose and variant objects allow users to apply dosing strategies and store alternative quantity values.
• Models can incorporate both fixed and random effects for population-level analyses.

Simulation Capabilities

• Simulations can be run using a variety of deterministic and stochastic solvers through the SimBiology Model Analyzer or via programmatic tools.
• Automatic unit conversion brings all model quantities into a consistent unit system.
• Reports can be generated directly from analyses.
• Monte Carlo simulations and parameter scans allow assessment of model behavior under different conditions and dosing regimens.
• Interactive sliders enable exploration of how quantity variation affects model response.

Parameter Estimation and Analysis

• Parameter estimation is supported using nonlinear regression with local, global, or hybrid methods.
• Parameter and prediction confidence intervals can be calculated.
• NLME modeling accounts for both fixed and random effects.
• Non-compartmental analysis (NCA) computes PK parameters from drug concentration time courses without requiring a compartmental model, and supports single or multiple dosing schemes.

Sensitivity Analysis

• SimBiology supports both global (Monte Carlo-based) and local (derivative-based) sensitivity analyses.
• Sensitivity analysis is used to explore the effects of variations in model quantities on model response and to identify key pathways and parameters.

Performance and Scalability

• Large models or Monte Carlo simulations can be accelerated by converting to compiled C code.
• Performance can be further improved by distributing computations across multiple cores, clusters, or cloud resources using the Parallel Computing Toolbox.

Sharing and Collaboration

• Apps can be created in App Designer, packaged with MATLAB Compiler, and hosted using MATLAB Web App Server.
• Collaborators can access and run these apps in a browser without installing any software.

Community and Extensibility

• SimBiology can be used programmatically with MATLAB scripts to customize and automate analyses.
• Community-contributed tools are available as add-ons, including gQSPSim (a MATLAB app for QSP model analysis), QSP Toolbox (workflow components for quantitative systems pharmacology), and VQMTools (software tools for QSP modeling).

SimBiology is available as a standalone product with a free trial option, and is also accessible to students through campus-wide MATLAB licenses. It is supported by extensive documentation, examples, video tutorials, and technical articles.