Posters
« Back
Heta language: Modularity and Reusability in QSP modeling platforms
EP39149
Poster Title: Heta language: Modularity and Reusability in QSP modeling platforms
Submitted on 17 Aug 2022
Author(s): Evgeny Metelkin
Affiliations: InSysBio
This poster was presented at ACoP12
Poster Views: 140
View poster »


Poster Information
Abstract: Objectives:
The size and complexity of mechanistic QSP models continue to increase. Such a process is an evident effect of the native complexity of biological structures. The expanding size is a challenging problem for efficient handling such models and modeling platforms. The most promising way to manage complexity is to decompose them into smaller modules which is easier to manage.
Modules can be linked to get the new modeling platforms or be parts of other modules providing modules hierarchy. This time-tested approach is widely used in the programming and engineering area but seems to be underestimated in the bio-math modeling area.
Heta is a modeling human-writable language (https://hetalang.github.io) for QSP and other biomathematical models supporting modularity [1]. This study reviews the features of the Heta language related to code modularity and reusability.

Methods:
Modularity in the Heta language is supported by the following important features. (1) Processdescription format as compared with ODE-based approach is more applicable for modularity because it
allows to separate dynamic values and processes as well as to add new components easily without rewriting ODEs. (2) A modeling platform can be split into several files and joined hierarchically with the “include” statement. (3) Heta compiler supports the different representations of model parts: Heta code, tabular format, JSON/YAML, and SBML. (4) The “namespace” statement supports the creation of several independent models in a single platform. (5) The “import” statement allows reusing components from one model in another one. (6) Any component or namespace can be cloned for the easy creation of repeated structures.

Results:
We applied the Heta-based modular approach for medium- and large-scale modeling platforms, including commercial and open-source modeling projects (https://github.com/insysbio) developed in InSysBio. Formalism simplifies the creation of complex models and allows using of shareable parts. The approach was also used in Immune Response Template (https://irt.insysbio.com) web application to create models based on selected components. The study includes a series of examples demonstrating the approach and Heta syntax for modules: compact format of repeated components, developing the platform from reused models including SBML modules, creation of template modules for pharmacokinetic models. Joined modules form a modeling platform which can be translated into popular simulation formats with Heta compiler v0.6 [2]: Simbiology, DBSolve, Mrgsolve, Matlab, Julia, and others

Conclusions:
The Heta language has reach capabilities to organize the hierarchical structure of modules. The modularity allows users to write compact and expressive code without repeating similar parts. The modeling platform can be created dynamically based on previously created parts including external SBML submodels.
Summary: The Heta language has reach capabilities to organize the hierarchical structure of modules. The modularity allows users to write compact and expressive code without repeating similar parts. The modeling platform can be created dynamically based on previously created parts including external SBML submodels.References: [1] E. Metelkin. "Heta" is a New Declarative Language to Define the Large-Scale Systems Pharmacology
and Systems Biology Models. American Conference on Pharmacometrics 10, October 2019
http://dx.doi.org/10.13140/RG.2.2.14881.35682
[2] E. Metelkin. "Heta compiler" is a framework for the development and management of Quantitative
Systems Pharmacology modeling platforms. American Conference on Pharmacometrics 11, November
2020 http://dx.doi.org/10.13140/RG.2.2.18555.13602
Report abuse »
Questions
Ask the author a question about this poster.
Ask a Question »

Creative Commons