« Back
HetaSimulator is an Open Source Simulation and Parameter Estimation Software for QSP Modeling
Poster Title: HetaSimulator is an Open Source Simulation and Parameter Estimation Software for QSP Modeling
Submitted on 23 Nov 2021
Author(s): Evgeny Metelkin
This poster was presented at ACoP12
Poster Views: 224
View poster »

Poster Information
Abstract: Objectives: QSP modeling requires a vast range of computational procedures including solving largescale differential equations systems, Monte-Carlo simulations (MC), parameter estimation, uncertainty quantification etc. These technical aspects of QSP modeling are developing rapidly both in terms of methodology (new modeling frameworks, standards for parameter estimation, etc.) as well as in terms of computational methods (automatic differentiation, parallel computing, etc.). Software, which could establish the linkage between emerging QSP frameworks and fast computational methods, is of utmost importance for the field.
HetaSimulator is a simulation environment for QSP modeling written in Julia. It utilizes unique Julia features to run computationally expensive procedures such as parameter estimation, identifiability and predictability analysis and multiple MC simulations. The main goal of HetaSimulator is to make these operations easily accessible for routine modeling workflow.

Methods: Internally HetaSimulator relies on Julia language, which allows code precompilation and simple framework for parallel simulations. The package utilizes features of SciML ecosystem, like access to 300+ ODE solvers, automatic differentiation, symbolic Jacobian derivation, etc.
Heta simulator was designed as a part of Heta project ( and together with Heta language and Heta compiler [1] forms a full-cycle modeling environment. It can be used as a simulation backend for Heta based models as well as an independent tool for differential equations based models.

Results: HetaSimulator allows users to run single and MC simulations with an ability to alter the state of a model with various event types, like dosage regime. It proposes an interface to define a parameter
estimation problem, which supports multi-conditional fitting, different experimental data inputs and distribution types. HetaSimulator can be integrated with the LikelihoodProfiler package [2] to perform
identifiability and predictability analysis. HetaSimulator is intensively used in both research and commercial projects of InSysBio team and it proved effective for large-scale models (> 1000 ODEs).
Summary: Conclusions: HetaSimulator can be used as a single platform for models validation in the QSP field. It
was released as an open-source software package ( with a
number of use-case models, which illustrate the ability of HetaSimulator to run all the stages of models
validation in the single framework
References: (1) E. Metelkin. "Heta compiler" is a framework for the development and management of Quantitative
Systems Pharmacology modeling platforms. American Conference on Pharmacometrics 11, November
(2) I. Borisov, E. Metelkin “Confidence intervals by constrained optimization—An algorithm and software
package for practical identifiability analysis in systems biology” PLoS Computational Biology
Report abuse »
Ask the author a question about this poster.
Ask a Question »

Creative Commons