Paper - 10's Abstract

Modeling communications between nodes in large-scale molecular interaction networks as information flows is useful for analyzing relationships between individual network components. Discovering causal genes and dysregulated pathways using network analysis based on information flow models is a very active, current topic of research. Recent research provides cubic order polynomial time solutions to the problem. Considering the huge size of interaction networks, calculating information flows can be costly, even with cubic order algorithms. An improvement to the computing time has been achieved by using the concept of approximation algorithms which are useful in solving large instances of problems requiring numerous resources. Proofs for the approximation factor and implementation results for the proposed algorithm are presented.