GTOM - Generalized Topological Overlaping Measure

GTOM quantifies higher-order topological overlap to measure pairwise interconnectedness in biological networks for gene and protein interaction analysis.

Key Features:

• Generalization of Topological Overlap: Extends the traditional topological overlap measure to consider neighborhoods beyond direct adjacency, enabling assessment of higher-order connectivity.
• Higher-Order Neighborhoods (m > 1): Counts the number of shared m-step neighbors between node pairs to capture functional relationships not evident from direct connections.
• Normalization: Scales the measure to a standardized range between 0 and 1 to facilitate comparison across networks and datasets.
• Sensitivity versus Specificity Trade-off: Allows adjustment of neighborhood order and thresholds to balance sensitivity and specificity when defining pairwise interconnectedness and detecting modules.

Scientific Applications:

• Module Detection: Enhances identification of network modules—groups of densely interconnected nodes—within gene and protein interaction networks.
• Gene Neighborhood Analysis: Supports detection of functionally related genes or proteins that share significant higher-order connections despite noisy or incomplete direct interactions.

Methodology:

For each node pair GTOM counts shared m-step neighbors, normalizes the count to the 0–1 range, and has been applied to yeast co-expression networks and fly protein interaction networks.