PerPAS: Topology-Based Single Sample Pathway Analysis Method.

Area of Science:

• Computational biology
• Cancer research
• Bioinformatics

Background:

• Targeted cancer therapies require understanding cancer progression and drug resistance pathways.
• Personalized medicine necessitates computational tools for analyzing limited sample sizes.
• Standard statistical methods are often unsuitable for single-sample analyses.

Purpose of the Study:

• To develop a novel pathway analysis method (PerPAS) for single-sample estimation of pathway activity.
• To identify significantly altered pathways and key nodes within them.
• To assess the utility of PerPAS in identifying patient survival-associated pathways in breast cancer.

Main Methods:

• PerPAS integrates pathway topology and transcriptomics data for single-sample pathway activity estimation.
• The method identifies altered pathways and key contributing nodes.
• Performance was evaluated using synthetic datasets and real breast cancer data.

Main Results:

• PerPAS successfully identified four pathways associated with patient survival in breast cancer.
• These findings were validated across three independent breast cancer cohorts.
• PerPAS demonstrated superior performance compared to two existing single-sample pathway analysis methods.

Conclusions:

• PerPAS is an effective tool for single-sample pathway analysis in cancer research.
• The method can identify biologically relevant pathways linked to patient outcomes.
• PerPAS offers a valuable approach for personalized medicine applications in oncology.