A multi-omics framework for survival mediation analysis of high-dimensional proteogenomic data

Area of Science:

• Biostatistics
• Genomics
• Bioinformatics

Background:

• Survival analysis is vital for time-to-event outcomes like disease progression.
• Current causal mediation methods for survival analysis often use Cox regression and overlook multi-omics data integration.
• This limits comprehensive understanding of complex biological pathways.

Purpose of the Study:

• To propose SMAHP, a novel method for survival mediation analysis.
• To simultaneously handle high-dimensional exposures and mediators, integrating multi-omics data.
• To identify causal pathways influencing survival outcomes using a robust statistical framework.

Main Methods:

• Introduction of the accelerated failure time (AFT) model within a multi-omics causal mediation framework.
• Simultaneous analysis of high-dimensional exposures and mediators.
• Application to proteogenomic data for head-and-neck carcinoma.

Main Results:

• SMAHP demonstrates high statistical power and effective false discovery rate (FDR) control in simulations.
• Identified a gene mediated by a protein influencing survival time in head-and-neck cancer.
• The method outperforms existing approaches in simulated scenarios.

Conclusions:

• SMAHP offers a robust framework for multi-omics survival mediation analysis.
• It advances the integration of omics data for discovering causal survival pathways.
• The R package is publicly available for broader research application.