SAMNetWeb

SAMNetWeb integrates mRNA expression and global proteomics across multiple experimental conditions to identify pathways activated in heterogeneous high-throughput datasets using a multi-commodity flow formulation on protein interaction networks.

Key Features:

• Multi-commodity flow algorithm: Employs a multi-commodity flow-based algorithm that enforces sharing of underlying protein interactions across experiments.
• Multi-omic integration: Simultaneously analyzes two distinct data types, such as mRNA expression and global proteomics, within the same analysis.
• Multi-condition pathway identification: Identifies pathways that are distinct to or shared among multiple experimental conditions.
• Support for heterogeneous datasets: Applies to data from genetic screens, mRNA expression assays, and global phospho-proteomic experiments.
• Pathway-centric functional enrichment: Pinpoints biological processes and pathways that best explain observed changes in high-throughput data.
• Network-level visualization: Maps identified pathways within a unified protein interaction network to relate processes across experiments.

Scientific Applications:

• Integrated pathway detection: Detection of pathways activated across combined mRNA expression and global proteomics datasets.
• Comparative pathway analysis: Comparative analysis to find pathways common or unique across treatments or experimental conditions.
• Analysis of screens and phospho-proteomics: Interpreting results from genetic screens and global phospho-proteomic experiments in the context of protein interaction networks.
• Noise-robust interpretation: Mitigating noise in high-throughput experimental systems by integrating diverse data types to highlight consistent biological signals.

Methodology:

Formulates integration across experiments as a multi-commodity flow problem on protein interaction networks that requires experiments to share underlying protein interactions to identify distinct and common pathways.