Our unique modeling platforms facilitate the rapid and intuitive development of models, virtual populations, and clinical trial simulations, using simple, maintainable processes that generate quick and reliable predictions of clinical outcomes

  • Thales Model Development Platform: A reaction-oriented, rapid development platform that automates and streamlines the QSP modeling process
  • Disease Specific Models: Validated models of diseases that can be used for clinical trial predictions and hypothesis testing. We can customize models for our customers.
  • Component Model Libraries: Modules that can be used as standalone models or as a base for a customized model. Customers can use them in their own models, or we can customize a model for them.

Thales Model Development Platform

Imagine a QSP environment where building models occurs rapidly, model simulation times are fast, and over-engineering models is unnecessary because models can be adapted rapidly to changing requirements


  • A model building platform that automates and streamlines the QSP modeling process
  • The user describes biological influences to model and provides data to constrain parameters
  • Thales generates models from influences and assists users with iteratively fitting virtual populations to data
  • Thales dramatically speeds the process from biology research to validated virtual population

A “Top Down” approach:

  • Know the specific use of the model
  • Start with simple, prototype models that support the desired behavior, and add complexity and refit as needed/when data is available to constrain
  • Encourage lightweight, iterative design over heavyweight fixed design: Models are only as complex as they need to be to represent the data
  • Model build cycle times are up to 10x faster
  • Model revisions are fast (add new biological influences – new model and population are output)
  • Quickly generate patients, assess fit and validation performance of candidate model structures, then refine model design

Model Goals:

Mathematical models must be assessed along many dimensions. Most obviously, a model must reproduce expected behaviors for known inputs. However, good fits alone are far from sufficient. A high quality model should:

  • Generalize well (predict behavior for experimental conditions it was not to fit to)
  • Minimize compute costs (run-time, memory, processor requirements)
  • Minimize numerical problems (non-convergence or other artifacts)
  • Minimize excessive/artificial sensitivity to model parameters or perturbations
  • Maximize parsimony – miss nothing essential, and include nothing non-essential
  • Minimize reliance upon poorly constrained mechanisms – model complexity should be limited by supply of available knowledge
  • Be easily readable, modifiable, and maintainable

Building models that simultaneously satisfy all of these criteria is difficult, time-consuming, and frustrating. This is often a discovery process that is poorly documented and difficult to reproduce. Thales aims to help streamline the process of obtaining such models.

We are actively developing Thales tools that expand upon our traditional Thales approach:

Computer Aided Model Design:

Model construction raises countless design questions about which mechanisms to model, how to abstract them, and what the consequences of different designs would be. Worse yet, a good model must strike a balance between satisfying many conflicting and sometimes subjective criteria. This results in an iterative process of implementation, tuning, simulation, and refinement.  The result can be a high quality-model, but at a great cost of time and effort.

We are currently developing the Thales Model Architect, which will provide tools to help automate this process. Consider two workflows:

Traditional Thales: A modeler makes all design decisions about which species and interactions will be modeled, and writes reactions that provide a definitive “blueprint” for the interactions in the model. Thales implements the model and fits it to user-provided data. After fitting and validating, the modeler weights the results of these runs, proposes updated blueprints, and repeats.

Thales Model Architect: A modeler writes a “meta-blueprint” that describes not a single design, but rather a menu of potential design decisions from which many differing blueprints could be generated. The Architect then efficiently explores the space of possible model designs, automatically implementing and fitting candidate models, and scoring the resulting models against all relevant requirements.

In addition to sparing modelers from the tedium of the design-fitting loop, the Model Architect will offer several other advantages:

  • Provides formal documentation of the design options that were entertained
  • Provides formal documentation of the criteria for judging the model-driven
  • Provides a documented, reproducible, and objective criterion for model selection
  • Frees users to explore more possibilities than would otherwise be considered, resulting in better understanding of model tradeoffs, and converging to a model that better suits the user’s needs

While the Model Architect is primarily intended for building new models from scratch, it can also be used as a form of model reduction to seek out more parsimonious simplifications of an existing model.

Computer Aided Model Diagnostic: 

Simply evaluating the quality of a given model design can be a challenge – even a good design is only useful when properly tuned.  When a model fails to reproduce desired behaviors during fitting and validation runs, time-consuming debugging is often required to determine the cause of the failures. Potential causes of failures are nearly endless:  flaws in model design, careless errors in model implementation, contradictory fitting targets, poor choice of parameters to fit, errors in target data, poor use of optimization algorithms, etc. Not only is this process painful and costly, but it also discourages users from experimenting with the model for fear of resuming the fitting-debugging cycle.

A recurring question for modelers is whether their design truly isn’t capable of producing the desired results, or whether the proper tuning of that model just hasn’t been found.

We are developing Thales tools to help automate the process of diagnosing model failures. Our vision is that when Thales fitting hits a roadblack, Thales can:

  • Explore modifying parameter bounds or distributions within user-specified constraints to see if there are artificial limitations preventing fits
  • Analyze patterns in failures to identify constraints which appear to be unsatisfiable with a given structure
  • Explore the space of possible behaviors that can be produced by subcomponents of the model, helping to isolate potential structural limitations on output
  • These features can be deployed in concert so that Thales is simultaneously exploring the best fit that can be obtained with a given structure, while entertaining the possibility of switching to a different model structure, and updating its belief that a particular model is fittable.


Aegis Model Development Platform

The Immunetrics Aegis™ integrated model development platform is available for licensing. We work with our collaborators to determine licensing options that makes sense for them.

Aegis™ is a Software as a Service (SaaS) platform which gives model developers a unique and efficient view into their models. By providing quick access to configuration options and run-time information, modelers can author, visualize, and provide multiple execution scenarios to their models with ease.

The true power of Aegis™ is that all of the model simulation is executed on Immunetrics’ highly parallelized processor grid, which allows the user to execute a large number of models in parallel without needing to have expensive, high powered computing resources. This is particularly useful when trying to find values for parameters which can not be observed in the real world, or when a large diverse ensemble of models is needed.

Aegis™ Platform

  • Removing the computational barrier of large models: In addition to solving models faster, our Simulation Engine enables modelers to build more maintainable and testable systems
  • Exploring large systems: On the foundation of our Simulation Engine, we built an entire suite of model analysis tools: Optimization, Sensitivity Analysis, Parameter Space Exploration, and more
  • Simulating clinical trials: We developed capabilities for large scale population generation for clinical trial prediction
  • Modeling tools and user interface: The Aegis frontend gives modelers easy access to the tools they need to create meaningful models

We offer a fully managed software as a service platform (SaaS) which allows execution of models on our highly parallelized grid of servers.

A full range of training and support comes with Aegis licensing:

  • On-site and remote training available
  • Customer service options
  • Access to Immunetrics’ large scale grid computing resource
  • Fixed licensing fee/year/user