TCGA-Assembler

TCGA-Assembler automates the download, assembly, and processing of public TCGA genomics, epigenomics, and proteomics data to produce integrated matrices for downstream integrative and comparative cancer analyses.

Key Features:

• Module A (data retrieval): Implements a recursive algorithm that leverages TCGA file-naming conventions to retrieve URLs and download public TCGA data across genomic features and cancer types.
• Module B (data processing and assembly): Provides preprocessing functions including mega-data assembly, data cleaning, and quantification of measurements to prepare data for integration.
• CombineMultiPlatformData: Creates comprehensive data matrices that align different types of genomic measurements across samples for integrative analysis.
• Implementation: Both modules are implemented as R scripts.
• Data-matching and reproducibility: Contains procedures to handle intricate data-matching steps caused by varying lab protocols and biotechnologies and can produce Firehose-type data via open-source scripts.

Scientific Applications:

• Integrative multi-omics analysis: Enables systematic investigation of cancer mechanisms across multiple molecular and regulatory layers by combining genomics, epigenomics, and proteomics data.
• Cross-cancer comparative studies: Facilitates comparative analyses across different cancer types using harmonized TCGA datasets.
• Method and tool development: Supports development and benchmarking of data-mining and analysis tools by providing reproducible, assembled TCGA datasets.

Methodology:

Implemented in R; Module A uses a recursive URL-retrieval algorithm that leverages TCGA naming conventions to download files, and Module B performs preprocessing, mega-data assembly, data cleaning, measurement quantification, and uses CombineMultiPlatformData to align and assemble multi-platform genomic measurements into comprehensive matrices.