PDGPCS

PDGPCS predicts personalized cancer driver genes by integrating patient-specific gene expression, gene mutation data, and pathway information using the Prize-Collecting Steiner Tree model.

Key Features:

• Integration of Personalized Data: PDGPCS constructs a personalized weighted gene interaction network by integrating patient-specific gene expression data with known gene/protein interactions.
• Incorporation of Edge Weight Information: PDGPCS incorporates personalized edge weight information within the gene interaction network to capture patient-specific interaction strengths.
• Prize-Collecting Steiner Tree Model: PDGPCS applies the Prize-Collecting Steiner Tree model to quantify the impact of each mutant gene on dysregulated pathways and aggregate influence scores across pathways.
• Validation on TCGA datasets: Experimental validation on four TCGA cancer datasets demonstrated improved performance compared with other personalized driver gene prediction methods.

Scientific Applications:

• Personalized medicine and therapy prioritization: Identification of individualized cancer drivers to inform diagnosis and targeted therapy strategies for individual patients.

Methodology:

A personalized weighted gene interaction network is built from patient-specific gene expression data and known gene/protein interactions; gene mutation data and pathway information are integrated and the Prize-Collecting Steiner Tree model is used to quantify the impact of each mutant gene on dysregulated pathways; mutant genes are ranked by aggregated impact scores across affected pathways.