HALD: A Human Aging and Longevity Knowledge Graph from PubMed Literature

A text mining-based biomedical knowledge graph constructed from all published literature related to human aging and longevity in PubMed. The dataset includes seven sets of files in JSON and CSV formats, such as literature metadata, entity and relation triples, and specific biomarker information. It was created by Zexu Wu and is available under an Open Access license.

Use Cases

• Extracting aging and longevity biomarker relationships based on the dedicated biomarker JSON files.
• Analyzing publication trends and impact in gerontology based on literature metadata including journal impact factors.
• Building or populating a Neo4j graph database for aging research using the provided Entities.csv and Roles.csv files.
• Identifying key entities and their aliases in aging literature based on the Entity_Info.json file.

Strengths

• Structured as a knowledge graph with explicit triples (source entity, relationship, target entity) in the Relation_Info.json file.
• Includes specific files dedicated to aging biomarkers and longevity biomarkers, facilitating targeted analysis.
• Literature metadata contains detailed fields such as PMID, journal impact factors, author affiliations, and publication dates.

Limitations

• Row count and total dataset size are unknown, which may limit suitability assessment for large-scale projects.
• Column-level documentation is absent; field semantics must be inferred from the description or after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

PubMed literature database.

Collection Method

Text mining of all published literature related to human aging and longevity.