FAIRification of Computational Models in Biology
1. This paper introduces a structured approach to assessing the FAIRness (Findability, Accessibility, Interoperability, Reusability) of computational models in biology, particularly those encoded in domain-specific standards like SBML and CellML within the COMBINE community.
2. The authors present the FAIR COMBINE Archive Indicators project, aimed at adapting the Research Data Alliance (RDA) FAIR indicators to the specific requirements of computational models and archives, creating a new set of 84 indicators covering both models and metadata.
3. Unlike previous efforts, this project focuses on developing a semi-automatic tool to assess FAIRness, integrating features such as automated metadata detection, visual interfaces, and improved handling of model and archive data.
4. The proposed assessment framework distinguishes between models and archives, providing tailored indicators for each. Archives are evaluated based on omex format standards, while models are assessed according to SBML, CellML, SBGN, and SED-ML standards.
5. The authors emphasize the importance of metadata, recommending the separation of data and metadata to enhance accessibility and reusability. They propose using unique IDs for each model, model version, metadata file, and archive to improve reproducibility.
6. Community-driven workshops and hackathons were employed to refine the indicators, engaging stakeholders from the COMBINE community, including developers, curators, and repository maintainers.
7. The tool prototype developed under this project provides automated assessment for some indicators and manual input options for others. The authors encourage further development of this tool for broader adoption and usability.
8. Recommendations include establishing community guidelines for FAIRification, improving metadata standards, and encouraging the use of the assessment tool as part of the model publication process.
9. The project highlights the role of the European Open Science Cloud (EOSC) and the Research Data Alliance (RDA) in promoting open science and improving the quality of computational biology research through enhanced FAIR practices.
@dagmarwaltemath @satagopam @kirubel_biruk @konigmatt @irinabalaur
💻Code:
github.com/FAIR-CA-indicator…
📜Paper:
biorxiv.org/content/10.1101/…
#FAIRification #ComputationalModels #Bioinformatics #FAIRPrinciples #COMBINE #OpenScience #Metadata #SBML #CellML #FAIRCOMBINE