CanDIG is built to be completely distributed. Each data provider handles their own data and users, with complete control over who can access each data set and how much, with federated analysis built on top of APIs to this data, so that data can be analyzed without being copied.
CanDIG is private by design, with differential privacy built into the project from the very beginning so that data sets can be analyzed without exposing sensitive individual data. It uses modern authentication and authorization technologies like OpenID Connect and UMA to provide secured API-based access to data.
By allowing researchers at participating sites to analyze national-scale data while keeping data private and under local control, we will enable new, Canadian-scale research programs, one-stop discovery of genomics data, and an "App Store" of interactive or long-running analyses.
Want to find out more about our architecture? Click Here.