iDASH is the only National Center for Biomedical Computing (NCBC) with an explicit focus on advancing the methods by which research data are anonymized, analyzed, and shared. This privacy-preserving infrastructure allows scientists to concentrate on advancing their research while knowing that their underlying infrastructure –– security, patient privacy, and data integrity –– is not compromised. To address such concerns, iDASH has built a biomedical cyberinfrastructure (BCI) on top of a HIPPA-compliant private cloud that enables the project to scale its calculations and storage as needed, while maintaining a high level of security.
Creating a secure technical infrastructure that directly enables biomedical computing is critical to the success of the current iDASH Driving Biological Projects (DBPs) underway. iDASH provides an intuitive, innovative portal for accessing algorithms, open-source software, data storage, and training to facilitate secure, patient-anonymous data analysis, and sharing. Moreover, it establishes a community-serving foundation for the future that is:
- Available: Redundant infrastructure throughout the design enhances availability and stability
- Collaborative: Fine-grained controls, privacy-preserving algorithms, user-friendly interfaces, and common data sharing environments support multi-investigator collaborations
- Extensible: Scalable storage, VM hosts, and networking as needed; provide separate, secure, development/test/production environments as well as a “sandbox” for more open development on non-PII data
- Flexible: Ever-changing requirements, tools, and technologies
- High-performance: Scalable 10 gigabit networking throughout ensures peak performance on high-throughput data such as next-generation sequencers
- Secure: Leverages proven, audited processes, and architectures to provide a HIPAA-compliant environment
The iDASH BCI provides a single, comprehensive set of facilities to develop new algorithms, open-source tools, computational infrastructure, and services to advance biomedical research. Developed as an open, community-serving, crowd-sourcing resource, the iDASH team is collaborating with biomedical, behavioral, and quantitative researchers to establish the nation’s most robust data repository for high-quality collections of data.
A model-based approach for the iDASH BCI enables stakeholder-driven engineering of the system. At its highest level, the foundation for the BCI includes data integration, data evolution, data sharing, workflows, computation, security, and policies. The iDASH BCI builds on previous work on Rich Services (RS), an architectural blueprint that promotes encapsulation, separation of concerns, reusability, and service-orientation, while enabling a direct and easy deployment. Thus, the BCI is constructed in a hierarchical fashion, simply by following the same blueprint when decomposing the constituent application or infrastructure services.
To ensure the integrity and security of the iDASH system, environments and zones are distinctly separated by firewalls. The three private environments – production, test, and development – enable testing and quality assurance of services, software, and tools to protect personally identifiable information (PII) data. The presentation, application, and data zones, each with its own specific access requirements and protections, support the needs of researchers in each zone. A separate and distinct “sandbox” offers an open environment for developers to evaluate “proof-of-concept” applications with non-PII.
The iDASH Confluence site is the point of reference for stakeholders and developers interested in contributing and sharing data, resources, and infrastructure. The JIRA system allows users and iDASH staff to monitor and track issues or report on a project; while Bamboo helps users build and improve critical functionality in their application.