Integrative Pathway-Level Survival Analysis in Breast Cancer Using ssGSEA and FDR-Corrected Cox Modeling of TCGA Cohorts

Overview

Breast cancer’s molecular complexity presents significant challenges in identifying robust and reproducible prognostic biomarkers. This study applies an integrative, pathway-level survival modeling framework to The Cancer Genome Atlas (TCGA) breast cancer cohort to uncover survival-associated biological processes using transcriptomic data. Individual patient pathway activities were quantified using single-sample Gene Set Enrichment Analysis (ssGSEA), enabling a sample-specific assessment of coordinated pathway dysregulation. These enrichment scores were evaluated for association with overall survival using univariate Cox proportional hazards regression, with multiple hypothesis testing controlled via the Benjamini–Hochberg false discovery rate (FDR). Four pathways remained statistically significant following FDR correction (FDR < 0.05). HALLMARK_PROTEIN_SECRETION (HR = 1.56, p = 0.000858, FDR = 0.015), HALLMARK_HYPOXIA (HR = 1.54, p = 0.000898, FDR = 0.015), and HALLMARK_UV_RESPONSE_DN (HR = 1.44, p = 0.000790, FDR = 0.015) emerged as the strongest adverse prognostic features. HALLMARK_TGF_BETA_SIGNALING (HR = 1.45, p = 0.002032, FDR = 0.025) also demonstrated a significant association with survival. These hazard ratios indicate that increased activation of specific biological programs is associated with elevated mortality risk per standard deviation increase in pathway activity. This work demonstrates that pathway-level modeling integrating ssGSEA with FDR-corrected Cox regression can identify biologically meaningful survival signals that may be obscured in gene-level analyses. By providing a scalable and reproducible computational framework for pathway prioritization, this study supports the application of systems-level transcriptomic modeling to inform prognostic research and precision oncology strategies in breast cancer.

Competition history

• CSEF 2026 Computational Science (Senior Division) · Entry S-07-17

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Computational Sciences projects
• CSEF Computational Sciences projects
• Every CSEF 2026 project