MPAC

MPAC integrates multi-omic data through prior knowledge from biological pathways to infer consensus activity levels for proteins and associated pathway entities, predict novel patient groups with distinct pathway profiles, and identify key pathway proteins with potential clinical associations.

Key Features:

• Multi-omic data integration: MPAC interprets DNA copy number alteration and RNA-seq data to determine genes' DNA and RNA states (repressed, normal, or activated) as input for pathway analysis.
• Inferred Pathway Level calculation: MPAC uses PARADIGM to calculate Inferred Pathway Levels (IPLs) from DNA and RNA states.
• Permutation testing: MPAC permutes DNA and RNA states to create a background distribution to filter IPLs and remove events observed by chance.
• Factor graph modeling: MPAC leverages network relationships encoded in pathways through a factor graph to infer consensus activity levels for proteins and associated pathway entities from multi-omic data.
• Patient subgroup prediction: MPAC groups biological samples by pathway activities to identify novel patient subgroups with distinct pathway profiles.
• Clinical relevance prioritization: MPAC identifies and prioritizes proteins with potential clinical relevance, such as associations with patient prognosis and immune cell composition.
• Downstream analysis and visualization: MPAC provides multiple methods for downstream analysis and visualization of pathway activity results.

Scientific Applications:

• Cancer subtype identification: MPAC predicted a patient subgroup related to immune responses in head and neck squamous cell carcinoma that was not identified by analysis with either DNA copy number alteration or RNA-seq data alone.
• Clinical outcome association: Key proteins identified via MPAC subgroup analysis have pathway activities related to clinical outcome as well as immune cell composition.
• Pathway protein discovery: MPAC enables identification of key pathway proteins with potential clinical associations for understanding cellular mechanisms in disease.

Methodology:

MPAC determines genes' DNA and RNA states (repressed, normal, or activated) from CNA and RNA-seq data, which serve as input for PARADIGM to calculate Inferred Pathway Levels (IPLs). MPAC permutes DNA and RNA states to create a background distribution to filter IPLs. MPAC leverages network relationships encoded in pathways through a factor graph to infer consensus activity levels for proteins and associated pathway entities from multi-omic data.