René Bastón, Executive Director Kathy McKeown, PI Coordinated by Columbia University
NE Regional Big Data Innovation Hub Provide frameworks for public-private, multi-sector collaborations to address high-priority challenges with data-driven solutions Columbia (PI)
By the Numbers Community/Communic ications: Events: Convened or participated in 28 events; over 6000 people Website Relaunch: Blog, Newsletter, Twitter; ~9000 visitors/month Altogether, 10s thousands reached
By the Numbers Add’l Fun Fundin ing and and In In-Kin ind Con ontr trib ibutio ions $4.9 million 10 New Proje ojects 3 Spoke Projects 4 Planning Projects Innovator Internships BD-Map - CRUX Cybersecurity Risk NTIS Joint Venture Partnership
Cross-Sector Outreach
Cross-Sector Outreach
Cross-Sector Outreach
Cross-Sector Outreach International Educational Data Mining Society
A Licensing Model and Ecosystem for Data Sharing
What’s Next?
Next Steps & New Directions • Expanded Access to Data • BD-Map Pilot Funded; initiative launch Q1 2018 • Collaboration/Consortia Models • Cybersecurity Risk – Seed funded; initiative launch Q1 2018 • Transportation – partnerships in the works • NTIS JVP • Non-profit
For More Information nebigdatahub.org rb70@columbia.edu contact@nebigdatahub.org @NEBigDataHub #BDHubs | #NEBigData tinyurl.com/NEBDHubList
Chirag Patel Noemie Elhadad, Vasant Honavar, Greg Cooper chirag@hms.harvard.edu @chiragjp www.chiragjpgroup.org
Integration of E and causal reasoning approaches for large-scale observational health research: key investigators
lower pollution higher pollution 15% increased risk for death Increase translational impact of this type of research through: scale and accessibility ! (the exposome , phenome , larger and more generalizable populations)
Many hypotheses that we need to address to understand relationship between disease risk and environment! How does socioeconomic context influence hospital use , disease rates , and recovery ? What is the effect of air pollution levels in disease ? Do adverse weather conditions influence hospital use ? What pharmaceutical drugs lead to adverse health outcomes ?
Integrating the ExposomeDB with OHDSI and causal modeling tools to drive and demonstrate discovery.
Capitalize on digitalized health record data (from around the world)! High-powered dataset(s) for discovery
… and where do we get environmental information?
Examples of sources of disparate environmental datasets available in the Exposome Data Warehouse Geological NASA - Cloud and Atmosphere Profiles NOAA Climate Data Pollution EPA Air Quality Surveillance Data Mart, or AirData , Soc io-Economic US Census American Community Survey (ACS) Epidemiological Chirag Lakhani CDC Wonder, USDA Food Atlas
Mashing up Exposome Data Warehouse with patient data from OHDSI PM2.5 home zipcode income f(location, time) encounter time Pollen count EPA AirData NOAA Climate American Community Survey Chirag Lakhani
ExposomeDB is ready to deploy! Team is writing a manuscript that describes the resource Re-usable Jupyter notebooks coming soon! Lakhani et al, in preparation
Causal discovery tools ready to go: ccd.pitt.edu
Nam Pho Please contact Chirag Patel for help or project ideas! http://chiragjpgroup.org/exposome-analytics-course
Examples of science enabled by these resources:
“What is the average PM2.5 in May 2016?”
What about drugs? Possible to repurpose existing drugs to alleviate disease risk?
Is bupropion associated with better glucose profiles before and after exposure to the drug?
Is bupropion associated with better glucose profiles before and after exposure to the drug? yes!
What’s next? • Execute more science! • Systematic ExposomeDB and causal inference tools integration with OHDSI • Disseminate ExposomeDB resources (Jupyter notebooks and server) • Host students!
Check http://nebigdatahub.org for project updates! Workshop/Hackathon (in New York or Boston, 2018) Cross-institution short internships Tutorials + Code + Data! Contact us for code and ExposomeDB data! chirag@hms.harvard.edu
Recommend
More recommend
