Multi-netclust

Multi-netclust identifies connected clusters in multi-parametric networks by integrating multiple network matrices using user-defined threshold values and applying a memory-efficient graph algorithm to detect clusters interconnected across all or any of the provided networks.

Key Features:

• Matrix Integration: Integrates multiple network matrices using user-defined threshold values to combine relationships across matrices.
• Graph Algorithm: Applies a memory- and time-efficient graph algorithm to identify connected clusters across the integrated networks.
• Performance: Scales to large networks, demonstrated on networks with over 10^6 nodes and 10^8 edges and processing them within minutes on standard computing hardware.
• Programming Language: Implemented in C/C++.
• Algorithm Efficiency: Designed for fast, memory-efficient operation suitable for large-scale network analysis.

Scientific Applications:

• Connected cluster extraction: Extraction of connected clusters from data represented by multiple network matrices in multi-parametric network analyses.

Methodology:

Integrates multiple matrices using user-defined thresholds and applies a memory-efficient graph algorithm to identify connected clusters; implemented in C/C++.