The Application Of Curvature In The Similarity Analysis For Biological Sequences

Based on the completeness of genome sequencing projects of the Human and the rapidheadway of Microorganism and Rice, more and more molecular enormous sequences datahave been generated. The need to analyze, process and store these data expedites thecombination of Molecular Biology, Mathematics and Computer Sciences, and created a newinterdisciplinary field-Computational Molecular Biology. It not only has great sciencesignificance, but also has huge economical benefit. It belongs to basic research, focusing onexploring the biological order of nature; also belongs to applied research, many of its researchfruits can be industrialized into products with high value immediately or soon. Recently,Computational Molecular Biology has been a very active research field in the Life Sciences.It is composed of Information Science, Computer Science, Life Science, Mathematics,Statistics, Physics, Chemistry and so on. It mainly focuses on complex computationsinvolving gene sequences, protein sequences by mathematical and computer science. Thecompare of biological sequences is the most basic issue, as for DNA sequences, even if weconsider a short segment of them, we can not obtain the whole information of the object itdepicts directly. However, if we compare different biological sequences, we can obtain someimportant information. But it's very complex, many issues are not solved yet. Anyway, thecompare and analysis of the biological sequences is one of the most basic and the mostimportant issues, it also has great significance to the research of life sciences. This thesismainly studied the similarity analysis of biological sequences. The main results obtained inthis thesis can be summarized as follows:We used the curvatures of curves smoothed by theβ-spline function to analyze thesimilarity of the DNA sequences for the first time and proposed the curvatures as a newinvariant. We analyze the similarities of the coding sequences of the first exon ofβ-globingene of 9 different species by our method. Our method is simple and has high veracity.