Abstract: A novel framework was presented for using evolutionary algorithm to identify the transcription factor binding sites from the noncoding region of the upstream of the coexpression genes. The application of the method in known promoter sequence sets which controled by the same transcription fact or showed that the proposed method can exactly find the binding sites with unigue conserved sequence. Compared with the classic method〖CD2〗Gibbs Sampler, the proposed method shows more effectiveness on finding the shorter binding sites.

Key words: genetic algorithms, particle swarm optimization, evolutionary computation