Research On Modeling And Simulation Method Of Biological Processes Based On Petri Net

It is very important to simulate biological processes in the post-genome era, which support the research on medicine and pharmacy forcefully. It is a difficult problem and a research focus to do the research on how to intuitively express biological processes recently.The thesis proposes Petri net as the basis of modeling biological processes because it can intuitively express dynamic nature phenomenon, according to features of biological processes themselves, obtained modeling method of Hybrid Petri net, Hybrid Functional Petri net and Stochastic Petri net through improving on Petri net.Because biological processes are very complex, it is difficult to express them intuitively and some typical processes are reused frequently, so the thesis proposes object-oriented Hybrid Functional Petri net. Encapsulate some complex reaction types as objects, including Elementary mechanisms, Enzymatic mechanisms and Gene expression mechanisms, simplify the description of complex biochemistry reactions; at the same time encapsulate typical biological processes, such as transcription and translation, so it can avoid a lot of repeated workload.As to the stochastic simulation of stochastic biological processes, if only use Stochastic Petri net, although it has standard graphics expression, which is isomorphic to Markov chain, along with the increase of models'scale and complexity, the number of states will increase exponentially, so it is very difficult to analyze models by the method which stochastic Petri net has; if only use stochastic algorithm, which has rapid simulation speed, but lack of intuitive graphical expression. So the thesis proposes that use stochastic Petri net as the expression of them in the front-end, use Gillespie-Gibson algorithm to perform the simulation processes in the back-end. Thus not only can express biological processes intuitively, but also obtain rapid simulation speed.Develop a stochastic simulation platform with the stochastic algorithm, and obtain good results after comparing with data achieved with other platform. As the experiment shows, it is feasible to use stochastic Petri net as the expression of biological processes in the front-end, and use Gillespie-Gibson algorithm to perform the simulation processes in the back-end.