Computational Model Of Breast Cancer Prognosis Based On The Interactions Between Cellular Signaling Pathways

As the most common malignant tumor in women worldwide and the second leading cause of cancer death after lung cancer, breast cancer is a serious threat to women’s life and health and has a lot of impact on the economy, society, family, and female psychology. The global morbidity of breast cancer has been on the rise since 1978 and was steady 123 per 100,000 in the recent years. Because of earlier detection through screening and increased awareness, as well as improved treatment, the mortality of breast cancer has been declining since about 1990 s.With the molecular biology gradually deeply applied to all aspects of the breast cancer research and the mass data generated by Human Genome Project, the study on molecular mechanisms of breast cancer prognosis also transformed from expensive clinical trials and evidence-based medical pattern to more perfect and individual medical pattern. The purpose of molecular biology is mainly to identify and their products, whereas the purpose of systems biology is to study biological system systematically. Therefor scientists introduced the knowledge of mathematics and computer science to the biological field and combined experimental data and simulated data to establish the computational model, which could reflect the interactions of biological system. Because of the complexity of the interactions between components of the biological system, systems biology model play an important role to both explore the relationship between the components qualitatively and quantify the effect of critical component on the system.Generally speaking, using systems biology model to simulate the biological system has the following advantages. Firstly, it can visually represent complex biological system in order to understand the interactions between components. Secondly, it can save time, manpower and material resources by using computer to simulate the experiment process. Thirdly, it can produce some data beyond the traditional experimental condition. Fourthly, it can predict the life phenomenon through interpret the experimental data reasonably and identify the integrity of biological systems and find the missing component. Fifthly, it can quantitatively explore the effect of each component on the whole system so as to identify the drug targets and drug doses.As to a kind of disease, doctors and patients usually care about the quality of prognosis. In the research of the prognosis of breast cancer, most of scientists were using experimental methods or statistical analysis. Especially the clinical treatment in the prognosis of breast cancer was always relied on doctors’ experience, and this therapy based on experience was usually lack of high accuracy and brought great pain and distress to patients. The improving understanding of the molecular mechanisms of the tumor growth process can improve clinical management of the disease and help doctors in clinical diagnosis and medication.In this paper, we established a computational model of breast cancer prognosis based on the interactions between proliferation, apoptosis, and immunity signaling pathways to study and analyze the interactions of proteins related in the prognosis of breast cancer. The main work includes the following aspects:(1) Overview of systems biology approachesIn this paper, firstly we systematically introduced the main contents and general process of systems biology, which is the foundation of our research. Systems biology approaches mainly includes: general composition and mechanism of signaling pathways, modeling methods of systems biology, parameters estimation and optimization, and parameters analysis and model analysis.(2) Construction of signaling pathways in prognosis of breast cancerFirstly we identified relevant signaling pathways of the development of breast cancer, including cell proliferation pathways and apoptosis pathways, in the KEGG database and BioCarta database. Then we added the effect of immune cells on the proliferation and apoptosis of tumor cell to more precisely simulate the tumor development in the prognosis of breast cancer. Finally, we integrated these trivial signaling pathways into a refined signaling pathway, which could systematically reflect the mechanisms between immunity, proliferation and apoptosis in breast cancer prognosis, based on the interactions between relevant proteins or cytokines.(3) Construction of computational model of breast cancer prognosisWe represented the signaling pathway using Petri net and meanwhile constructed the ordinary differential equations(ODEs) model of the signaling pathway by mass action equations. Using this way to construct the computational model of breast cancer prognosis could both describe the signaling pathway vividly and be used for calculation and analysis. To identify a suitable parameters estimation method for our model, we estimated the parameters by genetic algorithm(GA),particle swarm optimization(PSO) and simulated annealing(SA) algorithm. The result shows that PSO performs better in our model in the respect of fitting result, objective error and parameter error.(4) Parameter analysis and immune effect analysisBased on the constructed computational model of breast cancer prognosis, we did parameter analysis, including parameter identifiability analysis and parameter sensitivity analysis, and the result demonstrates the robustness of our model and the regulating role of RAF/MEK/MAPK signaling pathway in the growth of tumor. Then we did immune effect analysis to explore the effect of the immune factor EGF on the proliferation and apoptosis of tumor cell and the effect of immune factor FasL, Perforin and Granzyme on the apoptosis of tumor cell. These results could help doctors to improve the clinical management of the disease and drug targeting therapy.