TCGAnalyzeR

TCGAnalyzeR enables integrative visualization of pre-analyzed TCGA data, allowing users to create and compare subcohorts and visualize molecular and clinical results interactively for cohort and associated gene discovery.

Key Features:

• Simple Nucleotide Variation Analysis: Visualizes mutation data with driver prediction.
• Copy Number Alteration Analysis: Identifies and displays recurrent copy number alterations.
• Differential Expression Analysis: Compares tumor versus normal expression with pathway and survival analysis.
• Clinical Data Integration: Incorporates TCGA clinical data including metastasis and survival analysis.
• External Subcohort Integration: Includes subcohorts from the literature, curatedTCGAData, and BiocOncoTK R packages such as MSI, Immune, PAM50, Triple Negative, IDH1, miRNA, and metastasis subcohorts.
• Internal Patient Clustering: Determines patient clusters using iClusterPlus R package or signature-based expression analysis across five molecular data types.
• Custom Subcohort Creation: Allows users to create their own subcohorts and compare against existing external subcohorts and internal patient clusters.
• Interactive Visualization: Displays cohort-centric or gene-centric results interactively.

Scientific Applications:

• Pan-Cancer Molecular Profiling: Supports analysis of mutation, transcriptome profile, copy number variation, miRNA, and methylation data across all 33 TCGA cancer types.
• Subcohort Comparison: Enables comparison of user-defined subcohorts against literature-derived subcohorts and internally clustered patient groups.
• Gene-Centric Discovery: Facilitates interactive exploration of gene-specific patterns across multi-omics data types.
• Survival and Clinical Outcome Analysis: Integrates metastasis and survival data with molecular profiles for clinical pattern recognition.

Methodology:

TCGAnalyzeR re-analyzed five molecular data types (mutation, transcriptome profile, copy number variation, miRNA, and methylation) from ~11,000 cancer patients across 33 cancer types from TCGA. Internal patient clusters were determined using iClusterPlus R package or signature-based expression analysis. The platform integrates ~120 subcohorts from the literature, curatedTCGAData, and BiocOncoTK R packages.