RNA-KG v2.0: an RNA-centered Knowledge Graph with Properties

Area of Science:

• Biomedical Informatics
• Molecular Biology
• Bioinformatics

Background:

• RNA-KG is a biomedical knowledge graph for RNA interactions.
• Existing versions facilitate molecule classification and interaction prediction.
• There is a need for enhanced context-aware RNA data integration.

Purpose of the Study:

• Introduce RNA-KG v2.0, an upgraded biomedical knowledge graph for RNA molecules.
• Integrate manually curated interactions and detailed node attributes.
• Enable advanced, context-aware queries and link prediction for RNA research.

Main Methods:

• Integrated approximately 1.5 million manually curated interactions from 91 linked open data repositories and ontologies.
• Characterized relationships with standardized properties capturing experimental context (cell line, tissue, pathology).
• Enriched nodes with attributes like descriptions, synonyms, and molecular sequences from OBO, NCBI, RNAcentral, and Ensembl.

Main Results:

• RNA-KG v2.0 provides a comprehensive repository of RNA interactions with contextual information.
• The enhanced repository supports advanced queries considering experimental conditions.
• Integration with RNAcentral links RNA relationships with gene expression, structure, and pathways.

Conclusions:

• RNA-KG v2.0 significantly advances RNA research by providing a context-aware knowledge graph.
• The resource accelerates the discovery of novel therapeutic targets through improved data integration and analysis.
• Context-aware link prediction techniques are enabled, combining topological and semantic information.