nethet

nethet analyzes network heterogeneity in high-dimensional datasets to enable robust estimation and statistical comparison of network structures.

Key Features:

• High-dimensional network analysis: Provides methods for estimating and comparing network structures derived from high-dimensional data.
• Two-sample testing in Gaussian graphical models: Implements formal two-sample testing procedures for Gaussian graphical models, referencing Stadler and Mukherjee (2013, 2014).
• Network-based clustering algorithm: Includes a clustering algorithm tailored to identify distinct network clusters within heterogeneous datasets.
• Robust statistical methodology: Employs advanced statistical techniques for precise inference on network heterogeneity and structure.

Scientific Applications:

• Genomics and molecular biology: Analysis of high-throughput, high-dimensional genomic data to detect differences in molecular interaction networks across samples.
• Interdisciplinary network analysis: Comparative network analysis across scientific domains that require modeling of heterogeneity in network structures.

Methodology:

Implements two-sample testing within Gaussian graphical models and a network-based clustering algorithm for estimation and comparison of networks in high-dimensional data, following methodologies described by Stadler and Mukherjee (2013, 2014).