Research On Computer Modeling Of The Immune System

One of the significant features of the development of modern computer science and technology is that its interpenetration and intersection is getting closer with life science.The development of life science, especially the development of biology, has provided many new methods and new ideas for the development of computer science. At present, the research which aim at the cross field between life science and the computer science has become an important research direction in the field of scientific research, and achieved many remarkable results. As the second complex system of human body, its rich characteristics make it more and more popular among the computer scientists and researchers in engineering. Although the characteristics and mechanism of the immune system and immunology theory has been developed and applied in engineering field, the partial success of artificial immune system can not prove that the immune system is omnipotent, and it also cannot prove that the current theory of immune system are correct, both immunologists and the researchers of artificial immune system are lack of the understanding for many immune process and phenomena. Especially the latter one, many immune mechanism theory are controversial and even have not been found.Through the study of computer models for the immune system, we can further validated the immune mechanism on the controversial.It has great feasibility and application value.This thesis provides a new process for the research of the immune system.First, the immune system is a complex system, this thesis uses the theory of complex network related immune system for data mining. In this thesis, according to the related knowledge of statistics, such as node-degree、act-degree、clustering coefficient,we extract the cells of immune system. Finally, we find that in the innovative collaboration network the node-degree and act-degree are showing SPL distribution which means the interconnection between the cells and medium match the linear selection and the random selection both. In another word, the interactions of the cells secrete media also have different choices by different situations. By the way, through the network medium’s node-degree of distribution and act-degree of distribution, we can find the importance of various cells in the immune system,which subsequently provide a basis for establishing the conceptual model of a computer.Second, as the prototype of artificial life science -cellular automaton theory, this thesis based on the theory of cellular automaton model proposes an improved model of CS,using java programming language developed the simulate system of immune system. What’s more, we add influenza virus-related data, simulation and verify the reasonableness of the simulation system. In the end, we made a discussion of the system parameters.Third, although the immune system simulation model is a conceptual model, the simulation results from a certain extent, can show mutual relations and internal links between elements of the immune system inside as well. In recent years, the study of GEP in the world continue to heat up. We combine the advantages of GEP algorithm in terms of function mining with the simulation results of the immune system. At last, we obtain more satisfactory results.Overall, the immune system is a complex self-organizing, self-adaptive giant system. The way of using complex network for data mining, according to the mining data to construct the conceptual model and the simulation of objects, then with the help of GEP to mine function from the simulation results provides a good research method of medical research.