On Complexity Measures for Biological Sequences
CSB '04 Proceedings of the 2004 IEEE Computational Systems Bioinformatics Conference
| Hi-index | 0.00 |
In this paper, the role of sequence complexity in the construction of important nodes in protein-protein interaction (PPI) networks is investigated. We use two complexity measures, linguistic complexity and Shanon entropy, to measure the complexity of protein sequences. Three different datasets of yeast PPI networks are used to conclude the results. It has been shown that there are two important types of nodes in the PPI networks, which are hub and bottleneck nodes. It has been shown recently that hubs and bottlenecks tend to be essential in the process of evolution. Better understanding of the properties of these two types of nodes will shed light on why proteins interact with each other in the observed manner. We show that the sequence complexity of hubs are lower than that of non-hubs. But the difference is not significant in most cases. On the other hand, the sequence complexity of bottlenecks are lower than that of non-bottlenecks and the difference is significant in most cases. Modularity has an effective role in the construction of PPI networks. We find that there is no significant difference in the node complexity among different modules in a PPI network.