SignalingProfiler 2.0 a network-based approach to bridge multi-omics data to phenotypic hallmarks

Area of Science:

• Computational Biology
• Systems Biology
• Bioinformatics

Background:

• Understanding cellular signaling remodeling is key to disease mechanism research and drug target identification.
• Computational tools are essential for generating mechanistic hypotheses from complex multi-omics data.

Purpose of the Study:

• To introduce SignalingProfiler 2.0, an updated multi-step pipeline for generating mechanistic hypotheses on signaling events impacting cellular phenotypes.
• To provide a flexible and accessible tool for integrating and interpreting multi-omics data.

Main Methods:

• SignalingProfiler 2.0 integrates proteogenomic data with prior knowledge-causal networks to derive context-specific signaling networks.
• The pipeline employs statistical, footprint-based, and graph algorithms for data integration and interpretation.
• Benchmarking was performed on three proof-of-concept studies.

Main Results:

• SignalingProfiler 2.0 successfully generated hierarchical mechanistic networks.
• The tool recapitulated known and identified novel perturbed signaling pathways and phenotypic outcomes.
• The pipeline demonstrated efficacy in both human and mouse contexts.

Conclusions:

• SignalingProfiler 2.0 effectively addresses the need for deriving biologically relevant insights from complex multi-omics data.
• The tool's ability to extract interpretable networks facilitates a deeper understanding of cellular signaling.
• This updated pipeline supports advancements in disease research and therapeutic target discovery.