Based Support Vector Machine Prediction Of Regulatory Networks Within The Genome-wide Study

With the post-genome era, the current genome sequence for a biological understanding of the structure is still far from enough, it is also necessary to understand how genes are organized, the function of each gene is, what with the development of control and micro - the impact of environmental factors in a particular space-time domain spectrum to start its expression. In the cell and organism growth, secretion, cell lines, such as the process of orientation, the transcription level of gene expression regulation is a very important link in the evolution of transcription factors in the species plays an important role, and the performance of gene regulatory networks is a large number of transcription factor gene by the end of regulation and control of transcription and then translated into protein function in complex biological information, is to understand biological processes and an important aspect of gene function. The range of whole-genome transcription factor (transcription factor) to establish the regulatory networks has become especially important.Under normal circumstances, in the face of mass-like gene sequence data, and costly technology, complexity, low sensitivity, poor reproducibility of gene chip technology to meet the research is far from the status quo, when the use of machine learning methods for data on biological data Mining biological research has become a new method, the subject of the main use of support vector machine method, known transcription factor binding data Bioinformatics solution to genome-wide regulatory networks predicted the issue of the establishment.For the complexity of gene sequences using support vector machine method, known targets of yeast transcription factor and its research, originality of yeast genome sequence for the dimensionality reduction, The results show that the dimensionality reduction of the transcription factor after the data The expression can also be effective on the characteristics of transcription factor, which for other transcription factors found to be very helpful. Make use of existing experimental data reported in the literature information on the yeast genome-wide regulation for the prediction of the network. Sequence of the transcription factor through the effective dimensionality reduction, combined with support vector machine technology, we achieved a number of results than the algorithm for cluster analysis.