Network of tRNA Gene Sequences

We showed in this paper that similarity network can be used as an powerful tools to study the relationship of tRNA genes. We constructed a network of 3719 tRNA gene sequences using simplest alignment and studied its topology, degree distribution and clustering coefficient. It is found that the behavior of the network shift from fluctuated distribution to scale-free distribution when the similarity degree of the tRNA gene sequences increase. tRNA gene sequences with the same anticodon identity are more self-organized than the tRNA gene sequences with different anticodon identities and form local clusters in the network. An interesting finding in our studied is some vertices of the local cluster have a high connection with other local clusters, the probable reason is given. Moreover, a network constructed by the same number of random tRNA sequences is used to make comparisons. The relationships between properties of the tRNA similarity network and the characters of tRNA evolutionary history are discussed. PACS numbers: 89.75.HC, 87.23.-n ∗ [email protected] † [email protected] ‡ [email protected]