The Automated Construction and Verification of Physically Plausible Models of Physiological Systems

Computational modelling plays a central role in both academic research and clinical applications. As a result, ensuring model reliability has become increasingly important. Since the introduction of advanced computers in the 1960s, the engineering community has tried to increase the credibility of simulations by developing methods for verifying and validating mathematical modelling. However, despite reports on models violating fundamental physical and thermodynamic principles, no systematic approach exists for assessing model credibility and reliability in systems biology. Consequently, modellers in this field have introduced various methods to assess model outputs by comparing simulation results with experimental data. Additionally, some approaches constrain the probable states of a model through the evaluation of energy flux, thereby eliminating implausible states. Nevertheless, there is no comprehensive framework capable of evaluating a model's overall physical and thermodynamic plausibility. To address this, I developed a novel method to assess and verify systems biology models and demonstrated the utility of this method by implementing it in a Python tool. This tool implements checks for mass, charge, and energy conservation. It automatically reads CellML and SBML models and evaluates their consistency with physical and thermodynamic principles. When applied to the BioModels Database, my tool identified 182 models governed by mass-action kinetics. Among these, 22 were found to exhibit reversible reactions, and 13 complied with physical and thermodynamic principles, making them plausible. These results underscore the need for attention to physical and thermodynamic consistency in simulations of systems biology models. They also highlight that agreement with experimental observations is not sufficient for robust and reliable models.

Name thesis:degree_name

PhD

Level thesis:degree_level

Doctoral

Discipline thesis:degree_discipline

Bioengineering

Grantor dc:publisher

ResearchSpace@Auckland

Year dc:date.issued

2025

Author dc:creator

• Akbarpour Ghazani, Mehran

Advisors dc:contributor.advisor

• Pan, Michael
• Tran, Kenneth
• Nickerson, David P

• Verification and Validation
• Thermodynamic Consistency
• CellML
• SBML

Statement dc:rights

• Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Licence dc:rights.uri

• https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm

Handle dc:identifier.uri

https://hdl.handle.net/2292/76143

OAI identifier oai:identifier

oai:researchspace.auckland.ac.nz:2292/76143