TCGA Expedition: A Data Acquisition and Management System for TCGA Data

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• The Cancer Genome Atlas Project (TCGA) provides extensive genomic and clinical data for cancer research.
• Navigating and utilizing the large volume (1.2 PB) of TCGA data presents significant challenges for researchers.
• Existing public portals for TCGA data access have usability limitations.

Purpose of the Study:

• To develop a software solution, TCGA Expedition, to streamline the acquisition, management, and analysis of TCGA data.
• To create a robust infrastructure for computational cancer research utilizing TCGA resources.
• To facilitate the integration of TCGA data with electronic health records for enhanced research capabilities.

Main Methods:

• Developed TCGA Expedition software using Bash, Python, and Java for data download, harmonization, versioning, and storage.
• Implemented a participant- and sample-centered data directory with comprehensive metadata.
• Established the Pittsburgh Genome Resource Repository (PGRR) as a collaborative platform for data access and analysis.

Main Results:

• TCGA Expedition successfully created a versioned, local data repository with direct references to original and processed data files.
• The software enables flexible data searching via a web portal and provides user-centric data tracking and provenance tools.
• Locally accessible Whole Genome Sequencing (WGS) BAM files were made available, addressing critical challenges in data handling and processing.

Conclusions:

• TCGA Expedition is an open-source, freely available software solution for building local infrastructure to manage and analyze TCGA data.
• The software supports computational cancer research by simplifying access to large public genomic datasets.
• The developed repository enhances collaborative research by providing efficient data interrogation tools.