TENET

TENET characterizes targets in biochemical signalling networks by computing network topological features and applying support vector machine models to prioritize candidate targets.

Key Features:

• Topological Feature Computation: TENET computes a comprehensive set of topological features from signalling networks that capture node structural properties and connectivity.
• Support Vector Machine (SVM) Integration: TENET applies a support vector machine (SVM)-based approach to identify predictive topological features that discriminate known targets from non-target nodes.
• Characterization Model Generation: TENET generates a characterization model that specifies which topological features are crucial for distinguishing targets and how to combine them to quantify node target likelihood.
• Empirical Validation on BioModels: TENET was empirically evaluated using signalling networks sourced from BioModels and curated real-world outcomes, demonstrating improved performance relative to existing approaches.

Scientific Applications:

• Drug discovery: Enables prioritization of novel potential drug targets by integrating network connectivity and predictive topological features.
• Systems biology: Provides network-contextual characterization of nodes within signalling networks to inform mechanistic and functional analyses.

Methodology:

Extracting topological features from signalling networks; applying an SVM-based approach to discern predictive features associated with known targets; and developing a model that highlights significant topological characteristics for target discrimination and quantifies node likelihoods.