KRAKEN: A provenance-tracked knowledge graph for multiomic and wellness research
Existing general-purpose biomedical knowledge graphs tend to focus on disease mechanisms and drug repurposing, leaving multiomic and wellness-relevant content underrepresented. KRAKEN (Knowledge Research & Analysis Kit for Evidence Networks) addresses this gap by integrating existing graphs (including Translator KG Open, RTX-KG2, and ROBOKOP) with specialized sources such as RefMet, LIPID MAPS,…
KRAKEN, a knowledge graph designed for multiomic and wellness research, tackles the underrepresentation of such content in existing biomedical knowledge graphs. By integrating established graphs like Translator KG Open, RTX-KG2, and ROBOKOP, along with specialized sources like RefMet, LIPID MAPS, NIH Common Data Elements, and the Polygenic Score Catalog, KRAKEN offers a comprehensive resource.
With over 15 million nodes and 113 million edges spanning 62 entity types, the graph encompasses diverse data types. Adopting the Biolink Model as its semantic framework, KRAKEN ensures seamless compatibility with resources emerging from the NIH NCATS Biomedical Data Translator program. The graph's construction is streamlined by a modular build system, utilizing under 48 GB of memory during peak operations.
Users can tailor the graph's scope by selectively including or excluding sources, while built-in analytical tools facilitate multi-hop reasoning, subgraph extraction, text, vector, and hybrid entity search, as well as enrichment analyses. These features are accessible through an interactive web interface, a REST API, and a Model Context Protocol server, enabling direct consumption by agentic and LLM-based systems. KRAKEN is freely accessible at https://app.krakenkg.com.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
