Cross-Medicine Knowledge Graph for Type 2 Diabetes Drug Discovery

A biomedical knowledge graph constructed by Zekun Zhou, integrating multi-source data from Hetionet, SymMap, TCMBank, STRING, and TTD. It contains 15 entity types (245,235 entities) and 52 relation types (7,155,373 triples), covering 709 core Type 2 Diabetes Mellitus genes. The dataset was last updated on May 13, 2026.

Use Cases

• Prioritizing drug candidates based on graph embedding link predictions.
• Explaining drug mechanisms via a unified path scoring framework incorporating rule-based reasoning.
• Analyzing multi-target disease mechanisms such as insulin signaling sensitization and inflammatory regulation.

Strengths

• Integrates data from five distinct biomedical sources (Hetionet, SymMap, TCMBank, STRING, TTD).
• Contains 245,235 entities and 7,155,373 triples, providing a substantial graph structure.
• The ComplEx graph embedding model achieved a reported Hits@10 score of 0.48 for link prediction.

Limitations

• The dataset is very small at 5.5 KB, suggesting it may contain only metadata or a summary, not the full graph.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Integrated from Hetionet, SymMap, TCMBank, STRING, and TTD.

Collection Method

Entity alignment and relation consolidation using Jaccard and overlap-based fusion strategies.

Freshness

Last updated 2026-05-13 17:33:16; freshness should be verified.