FAIR Data Standards for AI in Plant Biology: Current Practice and Case Studies

Area of Science:

• Plant science
• Bioinformatics
• Computational biology

Background:

• Artificial Intelligence (AI) is crucial for analyzing complex plant data (molecular, phenotypic, environmental).
• Progress in AI for plant science is hindered by data structure, semantics, and interoperability issues, rather than algorithms.
• The Findable, Accessible, Interoperable, and Reusable (FAIR) principles are essential for effective data sharing and analysis.

Purpose of the Study:

• To review the current implementation of FAIR principles in AI-driven plant research.
• To introduce a readiness level perspective for assessing FAIR data maturity.
• To analyze how FAIR readiness impacts the feasibility and scope of AI applications in plant science.

Main Methods:

• Literature review of AI applications and FAIR data principles in plant science.
• Development of a readiness level framework for FAIR data, from unstructured data to FAIR Digital Objects.
• Case study analysis of molecular, phenotypic, and integrative plant science use cases.

Main Results:

• FAIR readiness is method- and context-dependent, influencing AI analysis feasibility.
• Task-focused AI can work with lower FAIR readiness levels.
• Integrative, transferable, and reproducible AI analyses require higher FAIR readiness, including rich metadata, semantics, and provenance.

Conclusions:

• A shift from focusing on FAIR compliance to building FAIR capability is needed for scalable AI-driven discovery in plant science.
• Community-driven infrastructure and practices are vital for accumulating curated and reusable data.
• Enhancing FAIR data maturity is critical for advancing AI applications in plant research.