The Reference Implementation, developed during the first IMPaCT-Data project, is a technical guide that details the components, documentation, and processes required to deploy a node of the IMPaCT-Data data analysis infrastructure locally, following the standards and good practices of the program. Its objective is to enable the integration and interoperability of clinical, genetic, and molecular data for research in precision medicine, in accordance with the guidelines established by the project.
In the Reference Implementation you can find both documents (such as recommendations on data and software) and general demonstrators as well as specific sections for each of the components of this first iteration of the infrastructure. Each of these sections includes a brief explanation of each component, the technical documentation needed to install and run the component, and the links needed to access each of them.
The Central Components are those needed to configure a central node that hosts the federation’s unique centralized services, while the Local Components are those that a partner needs to install locally to have access to the different tools and the federated data network itself.
Finally, the Evaluation section contains the explanation of and links to the component evaluation processes.
Testing of the authentication and analytics components including:
In this first demonstration, the functionality of federated user access and job execution through the use of shared computational resources is shown. The video shows a user connecting to the central node of the infrastructure using the credentials of one of the participating nodes, thus enabling authentication and access to the Galaxy platform. Following this, the user uses a tool with the data that they have stored in their personal cloud. The video shows how the execution of this tool is carried out in a distributed manner in several of the computing nodes of the IMPaCT-Data federated cloud, thanks to the Pulsar network. Finally, the user visualizes and has at their disposal the results in both HTML and raw format, as well as metadata and input and output parameters of each of the executions performed.
Testing of data access components:
In this demonstration, the cloud computing resource for biomedical data has a process to enable user access, including links to institutional accounts, and the use of passports and visas to control access to protected data. The video shows how a user connects to a tool stored at one of the nodes participating in the federated network. Initially this user does not have access to any datasets. Once the node administrator generates the corresponding visa, the user has access only to the dataset determined by that visa and no other. The video also shows how a different user, when entering the same tool, has access to a different dataset. This is because this second user has permission to access other data.