clipper

clipper identifies signal paths within biological pathway topologies to perform topological gene set analysis and link pathway-specific alterations to phenotypes.

Key Features:

• Graph-Based Representation: Clipper represents biological pathways as graphs with genes as nodes and interactions as edges.
• Empirical Two-Step Approach: Clipper applies an empirical two-step procedure of pathway selection followed by signal path identification.
• Pathway Selection via Statistical Tests: Clipper performs statistical tests on means and concentration matrices derived from pathway topologies to identify significant pathways.
• Signal Path Identification via Junction Tree Decomposition: Clipper decomposes pathway graphs into a junction tree to reconstruct signal paths most associated with a phenotype.
• Novel Algorithm for Topological Gene Set Analysis: Clipper implements a novel algorithm focused on identifying biologically significant signal paths within pathways.
• Validation on Simulated and Real Expression Datasets: Clipper has been evaluated on both simulated and real gene expression datasets for identification of coherent signal transduction paths.

Scientific Applications:

• Gene Expression Analysis: Clipper enables identification of pathway-localized signal changes underlying differential gene expression associated with phenotypes.
• Pathway Analysis: Clipper identifies specific signal transduction paths within pathways to study pathway alterations in complex biological systems and disease.

Methodology:

Represent pathways as graphs (genes as nodes, interactions as edges); compute means and concentration matrices from pathway topologies and apply statistical tests for pathway selection; decompose selected pathway graphs into a junction tree to reconstruct signal paths.