Abstract

In this paper we describe a novel method for extracting a set of nodes that best capture the connections between k given nodes of interest in a biochemical network. This method relies on the projection of the nodes of the network, seen as an undirected graph, into an euclidean space. Euclidean distances between nodes in the projected space correspond to their commute time distances in the original graph, a measure based on a random walk model on the graph. Commute time reflects the distance between two nodes while considering all paths connecting them. Results on artificial data illustrate the interest of this approach.

Citations