clipper

clipper is an innovative algorithm for gene set analysis using biological pathways, a widely employed statistical approach for gene expression analysis. Unlike existing methods focusing on pathway topology, clipper addresses the challenge of identifying specific portions of a pathway that are altered in response to a biological problem. The algorithm utilizes a two-step empirical approach based on graph decomposition into a junction tree to reconstruct the most relevant signal path within a pathway. In the first step, clipper selects significant pathways using statistical tests on the means and concentration matrices of graphs derived from pathway topologies. Subsequently, it identifies signal paths within these pathways that exhibit the most significant association with a specific phenotype.