The Prediction And Analysis Of Regulatory Motifs In Prokaryote

Bioinformatics has becoming a new interdisciplinary which began in 1990s. The gene and protein sequences of many kinds of organisms are sequencing by the advanced sequencing technology. As a result, there are many datas about bioinformatics which brings us many conveniences for the research of bioinformatics but also challenge for us to analyze these datas at the same time.The regulation of gene expression is a very complex biological processes, and gene regulatory network construction is also the focus and difficulty in the research of bioinformatics. The gene expression and regulation in Prokaryotic can be realized by the interaction between RNA polymerase and regulation factor which can regulate the expression of genes by binding the specific frag-ment in DNA sequences. These specific fragments usually are conservative, and we can call them regulatory motifs. The accurate prediction of Regula-tory motifs especially the transcription factor binding sites is the key to the accurate construct of gene regulation network model, we mainly analysis and predicted the regulatory motifs in prokaryotic in the paper.In this paper, we introduced the background and the meaning of research for prediction of regulatory motifs firstly, then introduced and analysed of the existing prediction algorithms of regulatory motifs. Based on these algorithms, we constructed a weighted graph and designed a new graph theory algorithm MDS for the prediction of regulatory motifs. This algorithm contains searching the maximal cliques and adding vertex to the cliques and refinement. Then, by the comparison between BOBRO,MEME, MotifClick algorithms and our algorithm, we knew that MDS can fast identify regulatory motif sequences and can ensure a certain accuracy. Our method is a good method.The innovation institute of this paper is:we designed a new graph theory algorithm MDS for the prediction of regulatory motifs. We can guarantee the accuracy of the final result and reduce the time of the prediction of regulatory motifs by the method of searching the maximal cliques and the principle of adding vertex in the algorithm MDS which can make us get more sequences that contains the real motif sequences and avoid the duplicate searching of some maximal cliques.