NeAT

NeAT analyzes biological networks, clusters, classes, and pathways to support structural and functional interpretation of protein interactions, regulatory pathways, metabolic processes, and high-throughput experimental data.

Key Features:

• Graph comparison: Compares two graphs to identify shared and distinct nodes and edges.
• Neighborhood exploration: Explores node neighborhoods to analyze local connectivity and neighbor relationships.
• Pathfinding algorithms: Computes paths between nodes using pathfinding algorithms.
• Graph randomization: Performs graph randomization techniques to assess the significance of network properties.
• Graph-based clustering: Performs graph-based clustering to identify network modules.
• Clique discovery: Detects cliques within networks to identify fully connected subgraphs.
• Cluster-to-network mapping: Maps clusters onto existing networks to contextualize cluster membership within network topology.
• Intersection detection: Detects intersections between different cluster or class types, including overlaps between co-expression clusters and functional gene classes.
• Supported data types: Handles protein interactions, regulatory pathways, metabolic processes, and high-throughput datasets such as yeast two-hybrid, mass spectrometry, and microarray experiments.
• Scalability and custom data: Processes large datasets and accepts custom data inputs for tailored analyses.

Scientific Applications:

• Structural network analysis: Enables structural analysis of protein interaction, regulatory, and metabolic networks.
• Functional module identification: Identifies network modules and cliques to reveal functional groupings.
• Integrative cluster–network analysis: Maps clusters to networks and detects overlaps to relate co-expression clusters with functional gene classes.
• High-throughput data interpretation: Applies network-based analyses to yeast two-hybrid, mass spectrometry, and microarray datasets.
• Statistical assessment of network features: Uses graph randomization to assess significance of observed network properties.

Methodology:

Implements graph comparison, neighborhood exploration, pathfinding, graph randomization, graph-based clustering, clique discovery, cluster-to-network mapping, and intersection detection.