GAC

GAC integrates high-dimensional genomic data and clinical variables to identify and visualize associations between gene expression and clinical endpoints using supervised principal component analysis.

Key Features:

• Supervised Principal Component Analysis (SuperPC): Extends principal component analysis by incorporating clinical outcomes to detect associations between genetic information and clinical endpoints.
• Clinical data type support: Handles continuous variables, binary outcomes, and time-to-event (survival) data.
• High-dimensional data integration: Integrates gene expression profiles and other genomic high-dimensional datasets with clinical data for association analysis.
• Visualization outputs: Summarizes association results using forest plots for binary and time-to-event analyses.
• Statistical analysis of clinical associations: Performs statistical analyses to infer potential clinical associations from high-dimensional genomic data.

Scientific Applications:

• Genomic–clinical association discovery: Identify associations between gene expression and clinical endpoints in high-dimensional datasets.
• Biomarker identification in oncology: Detect gene expression signatures associated with disease progression and treatment response in cancer cohorts.
• Survival analysis of gene expression: Relate expression profiles to time-to-event outcomes for prognostic studies.
• Support for personalized medicine research: Integrate genomic and clinical variables to inform hypotheses about patient-specific clinical associations.

Methodology:

Applies supervised principal component analysis (SuperPC), which extends PCA by incorporating clinical outcomes, to high-dimensional genomic data (e.g., gene expression) with continuous, binary, and time-to-event clinical variables, and summarizes results with forest plots for binary and time-to-event analyses.