primekg

You can query gene-disease associations in a knowledge graph by searching biomedical entities and retrieving their direct neighbors with relationship metadata to support drug repurposing workflows. This Skill extracts direct connected entities from the PrimeKG CSV knowledge graph to build actionable hypotheses.

Disease-centered context summarization in network pharmacology produces a compact overview of a disease's associated genes, drugs, phenotypes, and related diseases. It aggregates connected entities and relationship metadata from the knowledge graph to translate biological questions into evidence-linked relationships.

To find direct neighbors of a drug or gene in a biomedical knowledge graph, use node search with optional type filtering to locate the entity, then retrieve its direct connected entities with relationship metadata. This supports hypothesis building across gene-drug-disease-phenotype relationships.

Yes, this Skill uses pandas to analyze CSV knowledge graph data for precision medicine. It requires access to the PrimeKG CSV knowledge graph and leverages pandas to support node search, neighbor retrieval with optional relation filtering, and context aggregation across multiscale biomedical relationships.

This Skill provides limited MVP support for drug-disease path exploration by offering direct edge lookups and outlines for path-finding between two nodes. It applies to drug discovery and repurposing workflows by extracting direct associations and relationship metadata from the knowledge graph.

When querying phenotype-to-disease relationships, expect limitations in path-finding capabilities, as the Skill currently provides only direct edge lookups and outlines for path exploration between two nodes. It focuses on direct neighbor retrieval and disease-centered context summarization rather than complex multi-hop reasoning.