We used two state of art graph clustering algorithms, namely ClusterONE and Louvains modularity for the module detection. The Louvain method, in the first step, looks for small communities by optimizing modularity in a local way. In the second stage, it aggregates nodes of the same community and builds sellckchem a new network whose nodes are the communities. These steps are repeated iteratively until a maximum of modularity is reached. This process naturally leads to hierarchical decomposi tion of the network and results in several partitions. It measures the density of edges inside the community as compared to edges of inter communities and is defined as The cohesiveness of a cluster in ClusterONE is defined as follows where, Win denotes the total weight of edges within a group of vertices V, Wbound denotes the total weight of edges connecting this group to the rest of the graph while P V is the penalty term.

We used ClusterONE because of its ability to identify overlapping cohesive sub networks in weighted networks and was shown previously to detect meaningful local structures in various biological networks. We used the ClusterONE plug in available in Cytoscape for implementation. Results Analyses of known indications in disease drug network Starting with 1976 known indications from Kegg Medicus, we first filtered out diseases and drugs that do not have a known gene association in the Kegg database of disease genes and drug targets. This resulted in 1041 known indications representing 203 diseases and 588 drugs. Using this data, we found that of the 1041 known indications only 132 pairs share at least one common gene.

We then checked if any of the known indi cations share a pathway. To do this, we used the dis ease pathway and drug pathway annotations from Kegg Medicus. While this also revealed that only 116 disease drug pairs share a common pathway, what was surpris ing was that only 36 disease drug pairs share both a pathway and a gene. This demonstrates that disease drug relationships cannot be captured just through gene centric approaches. To analyze the characteristics of known indications further, we computed a distance measure between each of the known indication pairs in the human protein interactome was 1 if u v and 0 if otherwise and m 12 ijAij.

AV-951 Although the partitioning seems like an approximate method and nothing ensures that the global maximum of modularity is attained, several tests have shown that it provides a decomposition in communities with modu larity that is close to optimality. selleck products The implementa tion is available as a plug in in Gephi. We also used another graph clustering approach, ClusterONE , to find the disease drug modules. We calculated the shortest path for all known indications in the protein interactions network using JUNG. Of the 1041 known indications, we were able to compute the shortest paths for 1008 disease drug pairs.