Research On Graphical Representation And Its Application In Bioinformatics

Bioinformatics is a new cross subject which emerged in the late 80’s with the initiation of the human genome project. It is a subject which is closely related to human existence and life. With the rapid development of science and technology as well as the deteriorating living environment, more and more people are concerned about bioinformatics in today’s society, they attempt to find new ways to improve and solve the problems faced. In all methods, the study of mathematical models to show its essential features, the most laid a solid foundation for bioinformatics, which makes the research of bioinformatics researchers scratching their ideas forward with light. In this paper, the similarity of DNA sequences and brain regions relevance are analyzed by the graphical representation of this mathematical model.This paper constructs a novel combination of single-shot, on the basis of the establishment of a new 3D graphical representation of the sequence of DNA modelCircular Helix-like Curve(abbreviation CHC). The new 3D model had the advantages of simple and intuitive, no information is loss, can be used to characterize a single DNA sequence, it can also be used to identify the similarities and differences between DNA sequences. According to the geometric characteristics of CHC, the paper extracted a 12 dimensional vector from CHC as a mathematical description of the amount of CHC. Thus, the degree of similarity between sequences can be reflected by the Euclidean distance between vectors. Obtained the Euclidean distance similarity matrix by calculating, and then are given 11 species, 74 ribosomal RNAs, 48 HEV and 18 species of mammals in the phylogenetic tree by MEGA software, respectively. The four successful test results showed that: CHC is an effective tool for sequence analysis and comparison.Through the Human Brain Project research study, firstly the paper take the time series of brain areas f MRI data into a matrix with Pearson correlation coefficient; next construct a simple combination of a single radio unit circle is divided into 90 regions, exactly corresponds to 90 brain regions; then construct another combination of single-shot, this single shot make the correlation coefficient matrix mapped to the unit circle corresponding to 90 regions skillfully, and obtain a graphical representation model of individual brain regions. Based on graphical representation of compact, the paper extracted the geometric center vector from model as a mathematical description of the amount of individual brain regions. Finally, the distinction between mental illness patients and normal people was achieved by using k-means and neural networks clustering methods, the results obtained are satisfactory in whole.