Survival Genie

Survival Genie integrates single-cell RNA sequencing (scRNA-Seq) and diverse molecular inputs to perform survival analyses across cancer datasets for identification of prognostic markers and pathway-level associations.

Key Features:

• Single-Cell Data Integration: Integrates scRNA-Seq cluster markers and single-cell transcriptomics data for prognostic analysis.
• Diverse Molecular Inputs: Accepts gene sets, gene ratios, tumor-infiltrating immune cell proportions, gene expression profile scores, and tumor mutation burden as inputs for analysis.
• Extensive Dataset Collection: Contains 53 datasets covering 27 distinct malignancies from 11 cancer programs, including adult and pediatric cancers.
• Gene Expression Partitioning Methods: Supports partitioning by mean, median, quartile, and cutp to define expression-based risk groups.
• Comprehensive Analytical Outputs: Produces box plots for low and high-risk groups, Kaplan–Meier survival curves, univariate Cox proportional hazards models, and correlation analyses between immune cell enrichment and molecular profiles.
• Pathway and Gene Set Analysis: Performs pathway-level and gene set analyses, including gene ratio-based comparisons, to identify prognostic biomarkers and therapeutic targets.
• Canonical Cell Type Enrichment: Evaluates enrichment of canonical cell types across cancers to assess tumor microenvironment composition.

Scientific Applications:

• Integration of single-cell and bulk data: Combines single-cell transcriptomic signals with traditional genomic inputs to assess associations with clinical outcomes.
• Identification of prognostic signatures: Detects novel gene signatures and pathway-level associations linked to survival outcomes to inform targeted therapy development.
• Tumor heterogeneity and immune landscape characterization: Characterizes tumor heterogeneity and immune landscape dynamics by evaluating canonical cell type enrichment and immune–molecular correlations across cancers.

Methodology:

Performs Kaplan–Meier survival analysis and univariate Cox proportional hazards modeling, applies expression partitioning methods (mean, median, quartile, cutp), and conducts correlation analyses between immune cell enrichment and molecular profiles to relate molecular inputs to clinical outcomes.