HotNet2

HotNet2 identifies significantly mutated subnetworks of interacting genes in large-scale cancer somatic mutation datasets to help distinguish driver from passenger mutations and reveal functionally relevant pathways.

Key Features:

• Genome-scale gene interaction network analysis: Leverages a genome-scale gene interaction network to contextualize mutated genes and identify subnetworks with recurrent mutations across patient samples.
• Diffusion process for local neighborhood identification: Employs a diffusion process on the interaction network to define each mutated gene's local neighborhood or sphere of influence.
• Two-stage multiple hypothesis testing with FDR control: Implements a two-stage multiple hypothesis testing procedure to bound the false discovery rate for identified subnetworks.

Scientific Applications:

• Cancer genomics research: Identifies functionally relevant mutated gene subnetworks to study molecular mechanisms driving cancer.
• Somatic mutation analysis in glioblastoma and lung adenocarcinoma: Applied to glioblastoma and lung adenocarcinoma somatic mutation datasets, recovering established cancer-related pathways and identifying additional candidate pathways.
• Target and biomarker discovery: Prioritizes subnetworks that can suggest candidate therapeutic targets and biomarkers.

Methodology:

Uses a genome-scale gene interaction network, applies a diffusion process to map each mutated gene's local influence, identifies interacting gene subnetworks with recurrent mutations across samples, and validates subnetworks via a two-stage multiple hypothesis test that controls the false discovery rate.