GESUR

GESUR identifies transcriptional isoforms associated with patient survival outcomes to detect Transcripts Associated with Patient Prognosis (TAPPs) in cancer using data from The Cancer Genome Atlas (TCGA).

Key Features:

• Prognostic isoform identification: Identifies TAPPs—transcriptional isoforms associated with overall survival (OS) or disease-free survival (DFS) independently of gene-level associations, including principal isoforms that retain functional protein domains and isoforms that lack critical domains found in canonical gene isoforms.
• Survival analysis: Performs survival testing using the Log-rank (Mantel–Cox) test to assess associations between isoform or gene-set expression and patient survival outcomes.
• Customizable cohort thresholds: Supports adjustment of cohort thresholds to refine subgroup definitions for survival comparisons.
• Gene-pair analysis: Enables analysis using gene-pairs to evaluate combined prognostic effects.
• Cox proportional hazards: Reports Cox proportional hazard ratios with 95% confidence intervals for risk quantification in survival plots.
• TCGA integration: Leverages The Cancer Genome Atlas (TCGA) for large-scale identification and characterization of prognostic isoforms.

Scientific Applications:

• Isoform-level prognostic biomarker discovery: Detects isoforms whose expression correlates with OS or DFS for biomarker identification in cancer cohorts.
• Characterization of domain-altering isoforms: Highlights isoforms that alter protein domains and are enriched in known cancer driver genes to identify cancer-associated events.
• Study of isoform contributions to tumor progression: Enables investigation of how alternative isoforms contribute to patient prognosis and may inform translational research.

Methodology:

Uses data from The Cancer Genome Atlas (TCGA) to perform large-scale identification of TAPPs via survival analysis with the Log-rank (Mantel–Cox) test, supports gene-pair analyses and adjustable cohort thresholds, and reports Cox proportional hazard ratios with 95% confidence intervals.