PHONEMeS

PHONEMeS models signaling networks from untargeted phosphoproteomics mass spectrometry data by integrating kinase/phosphatase–substrate interactions to infer the flow of signal perturbations.

Key Features:

• Input data: Uses high-content shotgun untargeted phosphoproteomic mass spectrometry data as primary input.
• Interaction integration: Integrates kinase and phosphatase substrate–interaction information to connect phosphosites to upstream regulators.
• Network modeling: Constructs logical network models that represent propagation of signaling perturbations.
• Optimization formulation: Reformulated as an Integer Linear Program (ILP) to perform network inference.
• Pathway scope: Can model deregulated pathways both upstream and downstream of specified deregulated kinases, not limited to direct perturbation targets.
• Temporal analysis: Supports analysis of datasets with multiple time points to capture temporal propagation of signals.
• Biomedical applicability: Has been applied to diverse datasets of medical relevance to investigate phosphosignaling mechanisms and drug modes of action.

Scientific Applications:

• Signaling network reconstruction: Infer signaling network topology and directionality from phosphoproteomic perturbation data.
• Kinase/phosphatase activity inference: Link observed phosphosite changes to candidate upstream kinases and phosphatases.
• Pathway deregulation analysis: Explore upstream and downstream effects of deregulated kinases within signaling pathways.
• Temporal signaling studies: Analyze time-course phosphoproteomic data to map temporal propagation of signaling events.
• Drug mode-of-action investigation: Elucidate mechanisms of action of perturbations or therapeutics through inferred phosphosignaling networks.

Methodology:

PHONEMeS integrates kinase/phosphatase–substrate interaction data with shotgun untargeted phosphoproteomic measurements to construct logical network models and solves the resulting inference problem using an Integer Linear Program (ILP); it can incorporate multiple time points and model upstream/downstream deregulated kinase effects.