Home

Project Description

Background: Knowledge graphs (KGs) facilitate the representation of complex relationships among heterogeneous data types and have been used extensively in biomedical research to model biological phenomena. While many data-driven KG construction methods have been developed, they remain largely unable to:

Construct KGs from multiple disparate data sources
Combine KGs created by different systems
Collaborate or share KGs across institutions due to their inability to account for the use of different schemas, standards, and vocabularies.

Used extensively in life sciences research, the Semantic Web was created to resolve these types of knowledge integration problems. The Web Ontology Language (OWL) is a Semantic Web standard for a graph-based knowledge representation and reasoning framework. OWL is highly expressive, enabling the integration of heterogeneous data using explicit semantics, and allows for the generation of new knowledge using deductive logic. Unfortunately, existing OWL-based KG construction methods are often built using complicated programs or toolsets, in arcane or difficult to use programming languages and require extensive computational resources.

Solution: PheKnowLator (Phenotype Knowledge Translator), a fully automated Python 3 library explicitly designed for optimized construction of semantically-rich, large-scale biomedical KGs from complex heterogeneous data. The PheKnowLator framework provides detailed Jupyter Notebooks and scripts which greatly simplify KG construction, assisting even non-technical users through all steps of the build process. To accommodate a wide range of users and use cases, PheKnowLator has:

Three build types (partial, full, and post-reasoner)
Can include inverse edges to link nodes
Outputs KGs with and without OWL semantics (e.g. OWL-NETS)
Generates KGs in several formats (e.g. triple edge lists .txt, OWL API-formatted RDFXML, graph-pickled Networkx MultiDiGraph)

Collaborators

Translational Research Informatics Team


Bill Baumgartner	Ignacio Tripodi	Adrianne L. Stefanski	Jordan Wyrwa

Current Release ⚙

The resulting knowledge graphs and molecular mechanism embeddings are free to download and included as part of each release.

The current release is v2.0.0
Data for this release can be downloaded here

Publications and Presentations 📝📣📊

We presented this work (poster) at the 15th Annual Rocky Mountain Bioinformatics Conference held December 6-8th in Snowmass, Colorado.
Ignacio Tripodi will present results on MechSpy, a novel application that uses PheKnowLator to perform toxicological mechanistic inference at the 2019 meeting of The American Society for Cellular and Computational Toxicology held September 24th-26th in Gaithersburg, Maryland.
PheKnowLator is referenced in a review article on Knowledge-based Data Science in the biomedical domain.
PheKnowLator was mentioned in a recent blog post

Contact

We'd love to hear from you! To get in touch, please create an issue or send us an email 💌

License

This project is licensed under Apache License 2.0 - see the LICENSE.md file for details. If you intend to use any of the information on this Wiki, please provide the appropriate attribution by citing this repository:

@misc{callahan_tj_2019_3401436,
  author       = {Callahan, TJ},
  title        = {PheKnowLator},
  month        = mar,
  year         = 2019,
  doi          = {10.5281/zenodo.3401436},
  url          = {https://doi.org/10.5281/zenodo.3401436}
}