Submission Cycle Office Hours are back! This is a chance to learn about the submission cycle process and ask questions. Register for an upcoming session here.

Training: Accessing Shared Data

The Data Manager service, outlined in this training module, will be outdated soon. The web service has been deprecated.  For information on how to access S3 objects using Package Service, please view this GitHub repository.


The NDA holds and protects rich datasets (fastq, brain imaging, etc.) in object-based storage (Amazon S3). To facilitate access to these the NDA supports deployment of data packages (created through the standard NDA query tools) to cloud-hosted Oracle databases called miNDARs (mini-NDARs), and read-only access to associated data files such as omics or imaging data directly where they are stored in S3. By providing this access, the NDA envisions real-time computation against rich datasets that can be initiated without the need to download data. Direct download is also currently supported if necessary.


This tutorial series demonstrates the basic process to deploy a data package to a miNDAR, use it to obtain the S3 endpoints of data files, use your NDA account to generate temporary AWS credentials, and use those endpoints and credentials to access or copy objects in S3.

Loading training module...