Utility of network integrity methods in therapeutic target identification

摘要: Analysis of the biological gene networks involved in a disease may lead to identification therapeutic targets. It requires exploring network properties, particularly, importance individual genes. There are many measures that consider nodes and some shed light on significance potential optimality or set genes as This has been shown be case cancer therapy. A dilemma exists, however, finding best targets based analysis since optimal should highly influential in, but not toxic to, functioning entire network. In addition, therapeutics targeting single often result relapse compensatory, feedback redundancy loops offset activity associated with targeted gene. Thus, multiple reflecting parallel functional cascades simultaneously, require such We propose methodology exploits centrality statistics characterizing within is constructed from expression patterns both graph theory spectral theory. also origins topology, show how different available representations yield node results. apply our techniques tumor data suggest involving particular genes, pathways sub-networks an possible can facilitate individualized treatments. The proposed methods have identify candidate thought oncogenes nonetheless play important roles cancer-related pathway.