Systematically linking tranSMART, Galaxy and EGA for reusing human translational research data

Area of Science:

• Bioinformatics
• Data Science
• Genomics

Background:

• High-throughput molecular profiling generates complex data requiring advanced computational interpretation.
• Explosive data growth necessitates robust data management for efficient organization and integration.
• Existing systems often lack seamless integration between raw and interpreted data.

Purpose of the Study:

• To design and implement a data ecosystem for linking raw and interpreted molecular profiling data.
• To establish a framework for efficient data management and computational workflow execution.
• To facilitate data discovery, traceability, and reanalysis for clinical research.

Main Methods:

• Established an ELIXIR implementation study in collaboration with the Translational research IT (TraIT) programme.
• Utilized the European Genome-phenome Archive (EGA) for raw data storage.
• Employed tranSMART for interpreted and clinical data, and Galaxy for computational workflows.
• Integrated data by systematically linking repositories and using FAIR persistent identifiers.

Main Results:

• Developed a data ecosystem integrating EGA, tranSMART, and Galaxy.
• Successfully structured TraIT Cell Line Use Case (TraIT-CLUC) data for storage and cross-referencing.
• Enabled data flow from EGA to Galaxy for raw data reanalysis.
• Demonstrated user ability to select cohorts in tranSMART, trace to raw data, and perform reanalysis in Galaxy.

Conclusions:

• The proposed data ecosystem effectively links raw and interpreted molecular profiling data.
• FAIR persistent identifiers are crucial for stable data linkage and reuse across different data ontology levels.
• This approach enhances data discoverability, traceability, and reanalysis capabilities in clinical research.