RDBridge: a knowledge graph of rare diseases based on large-scale text mining.

Area of Science:

• Biomedical Informatics
• Genomics
• Drug Discovery

Background:

• Rare diseases impact 300 million globally, yet research faces hurdles like limited data and commercial interest.
• Challenges in rare disease research include data scarcity, complex relationship extraction, and data standardization issues for text mining.

Purpose of the Study:

• To develop a robust framework for rare disease data acquisition using text mining and knowledge graphs.
• To construct the most comprehensive rare disease knowledge graph, named Rare Disease Bridge (RDBridge).

Main Methods:

• Utilized text mining techniques to extract and integrate information.
• Developed a knowledge graph framework to consolidate diverse rare disease data.
• Built the Rare Disease Bridge (RDBridge) knowledge graph.

Main Results:

• Constructed RDBridge, the most extensive rare disease knowledge graph to date.
• RDBridge provides search functionalities for genes, potential drugs, pathways, literature, and medical imaging.
• The knowledge graph supports research into disease mechanisms, drug development, and clinical applications.

Conclusions:

• RDBridge offers a valuable resource for accelerating rare disease research and therapeutic development.
• The framework and knowledge graph facilitate data accessibility and integration for the rare disease community.