FAIRification of computational models in biology

Area of Science:

• Computational modeling
• Systems biology
• Scientific software engineering

Background:

• Computational models are crucial for understanding complex systems, especially in clinical applications.
• Quality assurance and transparency are paramount for clinical computational models.
• Effective communication of model capabilities is essential for their adoption and validation.

Purpose of the Study:

• To adapt existing Findability, Accessibility, Interoperability, and Reusability (FAIR) indicators for assessing computational models.
• To improve the transparency and communication of computational model features and capabilities.
• To guide the FAIRification process for models encoded in domain-specific standards.

Main Methods:

• Adaptation of the Research Data Alliance's FAIR indicators.
• Application of these indicators to assess computational models.
• Focus on models encoded in domain-specific standards, such as those from COMBINE ( a community for computational modeling in biology).

Main Results:

• Developed a framework for assessing computational models using adapted FAIR principles.
• Demonstrated how FAIR assessments can enhance model transparency and communication.
• Showcased the added value of FAIRification for computational models.

Conclusions:

• Adapted FAIR indicators provide a robust method for evaluating computational models.
• FAIR assessments facilitate better understanding and utilization of complex models.
• This approach is particularly beneficial for ensuring the quality and trustworthiness of models in clinical settings.