Comparing the biological coherence of network clusters

Detecting the modular structures from the protein-protein interaction network is important for understanding the organization, function and dynamics of a biological system. In order to identify functional neighbor- hoods based on network topology, many network

cluster identification algorithms have been devel- oped. However, each algorithm might dissect a network from a different aspect and may provide dif- ferent insight on the network partition. In order to objectively evaluate the performance of four com- monly used cluster detection algorithms: molecular complex detection (MCODE), NetworkBlast, shortest-distance clustering (SDC) and Girvan-Newman (G-N) algorithm, we compared the biological coherence of the network clusters found by these algorithms through a uniform evaluation framework. Each algorithm was utilized to find network clusters in two different protein-protein interaction net- works with various parameters. Comparison of the resulting network clusters indicates that clusters found by MCODE and SDC are of higher biological coherence than those by NetworkBlast and G-N algorithm. network cluster detection algorithms, biological relevance, function entropy, protein-protein interaction network Protein-protein interaction (PPI) networks are crucial for are denser than the connections to the rest of the net- [1] many biological functions . Almost every cellular work. These highly connected network clusters often process relies on transient or permanent physical bind- correspond to the basic molecular machineries in the ings of proteins. Currently, many experimental and cells. Proteins within each cluster are relatively homo- computational methods are available to detect or predict geneous in function, and the clusters are relatively inde- [2 ⚷ 9] [13] PPI in different organisms . The datasets thus gener- pendent of each other . Identifying such functional ated have provided us a chance to examine the biologi- neighborhoods from PPI network is essential for under- cal functions at the interactome network level. standing the functions, organization, dynamics and evo- The next challenge is to understand the biological lution of biological systems. [10] functional significance of the PPI networks . A great Several methods have been applied to the PPI net- amount of evidence has suggested that functional mod- work in order to detect modular structures. These algo- ules are cellular entities of complex biological sys- rithms differ from each other in their definitions of net- [1,11,12] tems . PPI network can be described as an undi- work clusters and in the clusters detected. Furthermore, rected graph whose nodes represent proteins and whose when different parameters are specified at the runtime, edges correspond to pair-wise interactions. Network Received February 15, 2007; accepted June 19, 2007 clusters are defined as sub-groups of PPI network whose doi: 10.1007/s11434-007-0454-z † Corresponding author (email: jdhan@genetics.ac.cn) proteins are closely linked and work together...

Website: hanlab.genetics.ac.cn | Filesize: -
No of Page(s): 7
Download Comparing the biological coherence of network clusters ... - Hanlab.pdf