PNC

PNC identifies personalized driver genes and models phenotype transitions between healthy and diseased states by constructing paired single-sample state transition networks and applying structure-based control using Feedback Vertex Sets (FVS).

Key Features:

• Structure-Based Network Control: Applies structure-based control principles from complex network theory to determine a minimal set of driver nodes required to steer a system from an initial state to a desired state.
• Personalized State Transition Networks: Constructs personalized state transition networks using a paired single-sample network construction method to capture phenotype transitions between healthy and diseased states from individual genetic data.
• Feedback Vertex Sets-Based Control: Implements a Feedback Vertex Sets (FVS)-based control method to identify personalized driver genes and manage unknown dynamics of individualized systems.

Scientific Applications:

• Cancer Research: Tested across 13 cancer datasets from The Cancer Genome Atlas (TCGA), identifying cancer driver genes enriched in established gold-standard lists and demonstrating superior performance compared to existing methods.
• Understanding Tumor Heterogeneity: Reveals personalized driver genes through network characteristics to provide insights into tumor heterogeneity relevant for developing targeted therapies.

Methodology:

Constructs personalized state transition networks from individual patient genetic data using a paired single-sample network construction method; applies a structure-based control approach based on Feedback Vertex Sets to pinpoint personalized driver genes.