rbsurv

rbsurv identifies genes associated with survival outcomes from high-dimensional microarray gene expression data by applying Cox proportional hazards modeling.

Key Features:

• Survival Modeling: Uses the partial likelihood of the Cox proportional hazards model to select genes associated with survival.
• Sample Separation for Robustness: Separates samples into training and validation sets to enhance robustness of gene selection.
• Iterative Forward Selection: Employs iterative forward selection to discover multiple distinct sets of survival-associated genes.
• Adjustment for Risk Factors: Enables inclusion of clinical risk factors as covariates in microarray survival modeling.

Scientific Applications:

• Prognostic Biomarker Discovery: Identification of genes whose expression correlates with patient survival for development of prognostic markers.
• Mechanistic and Therapeutic Insight: Use of survival-associated gene sets to inform disease mechanisms and nominate potential therapeutic targets or personalized treatment hypotheses.

Methodology:

Applies the partial likelihood of the Cox proportional hazards model for gene selection, uses iterative forward selection to generate multiple gene sets, separates samples into training and validation sets for robustness, and allows adjustment for clinical risk factors.