MOSClip

MOSClip identifies survival-associated gene modules by integrating multi-omic data—gene expression, DNA mutations, methylation patterns, and copy number variations—to improve survival analysis and reveal potential therapeutic targets in cancer.

Key Features:

• Multi-Omic Integration: Combines gene expression, DNA mutation, methylation pattern, and copy number variation data into a unified analytical framework.
• Module-Level Survival Analysis: Focuses on gene modules rather than individual genes to capture synergistic prognostic signals.
• Topological Pathway Analysis: Applies topological methods to identify pathways and network modules associated with survival outcomes.
• Dimensionality Reduction and Multivariate Modeling: Uses dimensionality reduction on multi-omic data followed by multivariate statistical models to test survival associations.
• Performance Evaluation: Assesses sensitivity and specificity using simulated data and demonstrates findings on The Cancer Genome Atlas (TCGA) ovarian cancer dataset.

Scientific Applications:

• Cancer Survival Analysis: Detects gene modules associated with patient prognosis across multi-omic profiles in cancer studies.
• Biomarker and Therapeutic Target Discovery: Identifies module-level biomarkers and candidate druggable targets linked to survival outcomes.
• Method Evaluation and Validation: Enables benchmarking of survival-association methods using simulated datasets and TCGA ovarian cancer data.

Methodology:

Data integration of diverse omic datasets; application of topological pathway analysis to discern pathways and modules; dimensionality reduction on multi-omic data; multivariate statistical testing to assess survival associations; evaluation using simulated data and the TCGA ovarian cancer dataset.