Identification Of Meiotic Recombination Spots Based On Phase-specific Sequence And Burrows Wheeler Transform

With the development of biotechnology and the implementation of organism genome projects, the focus of biological research shifts from the accumulation of data to the analysis and interpretation of these data transition, Bioinformatics(also known as computational molecular biology) emerges with the tide of the times. Its research area is very wide. For example, sequence comparison, phylogenetic analysis, protein structure prediction, Identification of essential and non essential genes, Meiotic recombination spots identification, predication of success for polymerase chain reaction and so on. In the thesis, we proposed a novel method to identify the recombination spots with the tools of mathematics and computer science.Identification of meiotic recombination cold/hot spots is critical for in-depth studying of the mechanism of recombination and the genome evolution process as well.In this study, by combining a local DNA structural property with the Burrows-Wheeler transform, a finite string is converted into a new one, which is called as the reduced sequence. By means of the reduced sequence for a phase-specific sequence, a 336-D feature vector is constructed to characterize a DNA primary sequence. A feature selection tool is used to rank these features, and then an optimum subset of features is built. On the basis of the optimal features, support vector machine(SVM) is employed to identify the recombination hot/cold spots. It is observed by the rigorous jackknife cross-validation test that the overall accuracy of our approach arrives at 84.92%, which is higher(from 0.83% to 4.53%) than that given by previous methods. This result suggests that the proposed approach is a promising tool for identifying the recombination spots.