A Modular Workflow for Model Building, Analysis, and Parameter Estimation in Systems Biology and Neuroscience.

Area of Science:

• Systems biology and computational neuroscience.
• Focus on biochemical pathway modeling at the molecular scale.

Background:

• Neuroscience research spans multiple scales, requiring diverse modeling tools.
• Interoperability and data standardization remain challenges in systems biology modeling.
• Existing tools often lack seamless integration for model building and analysis.

Purpose of the Study:

• To develop a workflow addressing model format conversion and tool interoperability issues.
• To adhere to FAIR data principles for enhanced model findability, accessibility, and reusability.
• To facilitate the building and analysis of biochemical pathway models.

Main Methods:

• Utilized the SBtab format for standardized storage of biochemical models and associated data.
• Implemented custom MATLAB scripts for parameter estimation and global sensitivity analysis.
• Developed a web-based application for biochemical model simulations with network-free and stochastic solvers, including geometry.
• Performed multiscale simulations in NEURON to demonstrate model convertibility.

Main Results:

• Established a workflow enabling smooth conversion between different model formats.
• Demonstrated the ability to simulate the same biochemical model in three distinct simulators.
• Successfully integrated a biochemical model into a biophysically detailed single neuron model via multiscale simulations.
• Enhanced the characterization of model properties through cross-simulator analysis.

Conclusions:

• The developed workflow significantly improves interoperability and data management for biochemical pathway models.
• This approach facilitates comprehensive model characterization by enabling simulations across multiple platforms.
• The methodology supports the FAIR data principles, promoting collaborative and reproducible research in systems biology and neuroscience.