RTX Reasoning Tool proof-of-concept (POC) software

The RTX Reasoning Tool POC software is being collaboratively developed by a team of investigators at Oregon State University, the Institute for Systems Biology, and Ohio State University.

Organization of this repository

subdirectory code

All software code files for the RTX proof-of-concept (POC) are accessible (and version-controlled) under this subdirectory. The RTX POC software code is 97% Python3, with the remainder being small bash scripts, Cypher queries, etc.

subdirectory data

Text data files for the RTX system that are deployed using git are stored under this subdirectory. There are only a few such files because the RTX POC software obtaines most of the information that makes up the biomedical knowledge graph by RESTfully querying web-based knowledge sources, rather than by loading flat files.

subdirectory team-private

This subdirectory contains notes and other documents for sharing within the RTX team.

License

The software that we would develop if awarded OT2 research support to build RTX would be made publicly available, under the MIT open-source software license, in the RTX GitHub area.

Dependencies outside of GitHub

• The RTX POC software depends on a SQLite cache file, orangeboard.sqlite. The POC software can build this cache file if the cache file is not pre-installed by the user; however, the POC software will build the knowledge graph much more rapidly the first time, if our pre-built cache file (63 GB) is used. The pre-built cache file is stored in Amazon S3 (link; requires pre-arrangement for access).

• Neo4j and Python modules: please see the README file in the code subdirectory for more information.

Issue tracking

Our team makes extensive use of the GitHub Issues system for project management. Browsing the open and (~100) closed RTX issues will provide a sense of our team's workflow and organizational style, at least for a one-month POC project.

RTX team's institutional affiliations

• Oregon State University (Lead Proposing Institution)
• Institute for Systems Biology (Subaward #1)
• Ohio State University (Subaward #2)

RTX leadership team

Name	Role	Email	GitHub username	Areas of relevant expertise
Stephen Ramsey	Project PI, Oregon State University	[email protected]	saramsey	compbio, systems biology
David Koslicki	Project Co-PI, Oregon State University	[email protected]	dkoslicki	compbio, graph algorithms
Eric Deutsch	PI, subaward, Institute for Systems Biology	[email protected]	edeutsch	bioinformatics, data management
Arnab Nandi	PI, subaward, Ohio State University	[email protected]	arnabdotorg	UI for querying knowledge-bases

RTX other key personnel

Name	Affiliation	Email	GitHub username	Areas of relevant expertise
Jared Roach	Institute for Systems Biology	[email protected]		genomics, genetics, medicine, systems biology
Arash Termehchy	Oregon State University	[email protected]	arashtermehchy	databases, knowledge graphs
Theo Knijnenburg	Institute for Systems Biology	[email protected]	tknijnen	systems biology, Translator
Sui Huang MD PhD	Institute for Systems Biology	[email protected]		medicine, systems biology
Gustavo Glusman	Institute for Systems Biology	[email protected]		computational genomics