ROBOKOP

reasoning over biomedical objects linked in knowledge oriented pathways

ROBOKOP is an open-source, modular, knowledge graph (KG)–based system that includes several key components: a biomedical KG; a user interface and associated question-builder tool; a collection of harmonized and interoperable knowledge sources, represented as KGs in a service called Automat and accessible via application programming interfaces; and a variety of supporting resources, including reasoning algorithms and tools to support deep exploration of the ROBOKOP KG and the Automat KGs. The ROBOKOP question-building tool allows users to construct queries, submit them to ROBOKOP, and receive scored-and-ranked answers that can be dynamically explored. The ROBOKOP Automat allows for programmatic access to the ROBOKOP KG and the Automat collection of harmonized biomedical KGs.

Ask a Question

Use the ROBOKOP Question Builder to construct a new query, then use the visualization tool to explore relevant publications.

Explore the Data

Learn about the data in ROBOKOP and how to access it programmatically using the Automat API.

Additional Tools

ROBOKOP KG

ROBOKOP KG is an open-source biomedical knowledge graph containing ~10M nodes and ~250M edges drawn from ~30 biological data sources and bio-ontologies. Query the ROBOKOP KG directly in the Neo4j browser.

ExEmPLAR

ExEmPLAR is a generic Neo4j query-and-answer exploration tool that supports queries of the ROBOKOP KG and other Neo4j databases.

ExEmPLAR is under development and should be considered an experimental alternative to the ROBOKOP UI.

RoboDocumentation

A collection of notebooks and other documents to help users get the most out of ROBOKOP.

Citations

If you use ROBOKOP in your work, please cite the following papers:

Bizon C, Cox S, Balhoff J, Kebede Y, Wang P, Morton K, Fecho K, Tropsha A. ROBOKOP KG and KGB: integrated knowledge graphs from federated sources. J Chem Inf Model 2019 Dec 23;59(12):4968–4973. doi: 10.1021/acs.jcim.9b00683. https://pubmed.ncbi.nlm.nih.gov/31769676/.
Morton K, Wang P, Bizon C, Cox S, Balhoff J, Kebede Y, Fecho K, Tropsha A. ROBOKOP: an abstraction layer and user interface for knowledge graphs to support question answering. Bioinformatics 2019;pii:btz604. doi: 10.1093/bioinformatics/btz604. https://pubmed.ncbi.nlm.nih.gov/31410449/.

Funding

ROBOKOP is a joint creation of the Renaissance Computing Institute (RENCI) at the University of North Carolina at Chapel Hill and CoVar LLC. The prototype was developed with funding from the National Center for Advancing Translational Sciences, National Institutes of Health (award #OT2TR002514). ROBOKOP's continued development is supported with joint funding from the National Institute of Environmental Health Sciences and the Office of Data Science Strategy within the National Institutes of Health (award #U24ES035214).

License

ROBOKOP is available under the MIT license.

Events

ROBOKOP has weekly office hours from 2-3pm EST. View details on our calendar or subscribe with the ICS file.