An approach to identify candidate compounds for treating Parkinson’s Disease using techniques rooted in network science combined with modern graph query languages to identify compounds (and genes) that are maximally associated with PD in the network
Through the work of the Integrative Data Science Lab (IDSL), we are pioneering new ways to rapidly improve disease treatment and drug discovery using integrative knowledge graphs and advanced machine learning approaches to profile and predict the biological effects of potential new drugs. We developed Chem2Bio2RDF, the first large scale linked public data graph for preclinical drug discovery; novel link prediction and data mining algorithms for finding hidden insights in large heterogeneous data graphs; and in our 2012 Drug Discovery Today paper laid out a strategy for using linked data and graph analytics to expand beyond the current single-target drug discovery model.
Current projects include researching knowledge graphs that encode computable networks for multi-mechanism complex diseases, including the PRIDE project (Parkinson’s Research through Integrative Data Experiments).
We are thankful to NIH NCATS, Indiana CTSI, the OpenPHACTS foundation, Eli Lilly, and Pfizer for funding of this work. Applications in this area are being commercialized in our company Data2Discovery Inc.
SELECT papers
These are a selection of our papers that we think are a good starting point, with links to the PDFs of the articles. For a full list of publications relating to knowledge graphs, see David’s Google Scholar page. Please contact David if you have trouble accessing any papers of interest.
RELATED BLOG POSTS
