簡易檢索 / 詳目顯示

回結果列表
研究生： 呂奇璋
Chi-chang Lu
論文名稱： 轉錄網路的平行演算模擬
Transcriptional Network Simulation Using Parallel Computation
指導教授： 林保宏
Pao-hung Lin
口試委員: 黃忠偉
Allen Jong-Woei Whang
孫家偉
Chia-Wei Sun
學位類別： 碩士
Master
系所名稱： 電資學院 - 電子工程系
Department of Electronic and Computer Engineering
論文出版年： 2014
畢業學年度： 102
語文別： 英文
論文頁數： 133
中文關鍵詞： 基因蛋白網路 、代謝網路 、轉錄網路 、系統生物學 、GPU 、圖形處理器 、圖形處理單元 、平行處理
外文關鍵詞： Gene protein network, metabolite network, Transcription network, System biology, GPU, Graphical processing unit
相關次數： 點閱：210  下載：2 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 轉錄網路(transcriptional network)已經漸漸被用來表現各種生物的模型，從微生物的模型，甚至到人類的模型。既使如大腸桿菌(Escherichia coli)那麼小的一種生物，其用來表現生物的轉錄網路通常有上千個網路“節點(node)”。所謂的網路節點則可能是信號、分子、基因或蛋白質。如果沒有計算機輔助，人類要了解網路的行為是有其極大的困難的。網路節點之間的相互關係用“邊 (edge) ”來表示，實際則進行分子層級的動力學，並且可以用微分方程來表示。一個網路模型就有可能是一個大型的聯立微分方程式系統。節點及邊所形成網路拓墣，有些在轉錄網路中出現的機率比較高的被定為網路模體(motif)。如上述的網路模體、節點、邊等各種的網路元素是在轉錄網路中的反映單元。一個轉錄網路可以視為是各種的網路元素的組合。因此產生的問題是轉錄網路中的各種的網路元素是否能夠適當的分配到平行演算法的其中一個層級，是否可以用平行演算法來模擬。本研究的目標是用平行計算的方法工具來求解轉錄網路的聯立微分方程式系統。本研究中所使用的語言Matlab™的高階程式語言。經由生命期及癌細胞二個轉錄網路的模擬，使用多個中央處理器(Central Processing Unit)進行的平行計算是產生較佳效率表現的較好選項。

    Transcriptional network has been increasingly utilized to model various organisms, from micro bacteria to human. A network model representing an organism, even as small as Escherichia coli, is typically large and complex, and usually includes thousands of nodes. Nodes may be signals, molecules, genes or proteins. Understanding the behavior of transcriptional network can be difficult for human without the aid of computer techniques. Interactions between nodes are represented by edges which carries molecular scale kinetic dynamics modeled by differential equations. A network model can be a large system of differential equations. Some topology of these node and edge in transcriptional network having high possibility of reoccurrence are identified as motif. Various network elements, such as motif, nodes, and edges are units of reactant in a transcriptional network. A transcriptional network can be interpreted as composition of various network units. Here raises a question whether these various kinds of network units in the transcriptional network can be modeled by parallel computing algorithm when well fitted to a level of parallel computing. The aim of the study is to solve the system of differential equations of transcriptional network using parallel computation utilities. Simulation in research is implemented in Matlab™ high level programming language. Through simulations of a life span network and a tumor cell network, parallel computing with multi CPU cores is recognized to a better option that yields better performance in simulating the transcriptional network in the research.

    1 Background 9 2 Introduction to system biology 10 3 Introduction to transcriptional network 11 4 Parallel computation utilities 15 5 Simulation of transcriptional network 18 5.1 Problem 1: lifespan network 18 5.2 Method 21 5.3 System specifications of a simulation computer 21 5.4 Simulation programming 23 5.5 Optimization of Euler method by GPU 28 5.5.1 GPU instructions 28 5.5.2 Euler Method 29 5.5.3 Euler method optimized by GPU 31 5.6 Result 41 5.6.1 CPU vs CPU multi-cores 43 5.6.2 CPU vs GPU 47 5.7 Problem 2: Tumor Network 49 5.8 Method 50 5.9 Simulation programming 50 5.9.1 Option 1: Euler sequential 52 5.9.2 Option 2: Euler parallel in CPU 52 5.9.3 Option 3: Euler GPU optimized 52 5.10 Result 53 5.10.1 CPU vs CPU multi-cores 53 5.10.2 CPU vs GPU 55 6 DISCUSSION 57 7 CONCLUSION 59 REFERENCE 60 Appendix A: Detailed description of Life_simulator.m 61 Option 1: Runge-Kutta method in sequential execution 61 Option 2: Runge-Kutta method in parallel execution by mult-cores 63 Option 3: Runge-Kutta method in parallel execution by GPU 64 Option 4: Euler method in sequential execution 66 Option 5: Euler method in parallel execution by mult-cores 68 Option 6: Euler method in parallel execution by GPU 69 Appendix B: Source code of pushbutton2_ButtonDownFcn() in Life_Simulation.m 72 Appendix C: Source code of pushbutton3_ButtonDownFcn() in Life_Simulation.m 73 Appendix D: Source code of Life_Simulator.m 84 Appendix E: Source code of pushbutton2_ButtonDownFcn() in Tumor_Simulation.m 97 Appendix F: Source code of pushbutton3_ButtonDownFcn() in Tumor_Simulation.m 98 Appendix G: Source code of Tumor_Simulator.m 99 Appendix H: Source code of TumorEulerODEsolverGPU06072.m 100 Appendix I: Source code of func.m 109 Appendix J: Source code of funcs.m 113

    1. Alon, U. (2006). An Introduction to Systems Biology: Design Principles of Biological Circuits. Israel: Chapman & Hall.
    2. Amir Abdollahi, C. S. (2007, 6 15). Transcriptional network governing the angiogenic switch in human pancreatic cancer. PNAS , p. vol. 104 no. 31.
    3. bio-physics-wiki. (2014, 6 7). Bacterial Growth. Retrieved 6 7, 2014, from bio-physics-wiki: http://www.bio-physics.at/wiki/index.php?title=Bacterial_Growth
    4. bio-physics-wiki. (2014, 6 7). bio-physics-wiki. Retrieved 6 7, 2014, from bio-physics-wiki.
    5. Heterogeneous computing. (2014, 6 7). Retrieved 6 7, 2014, from Wikipedia.
    6. Irina Voineagu, X. W. (2011, 6 16). Transcriptomic analysis of autistic brain reveals convergent molecular pathology. nature , pp. 474, 380–384.
    7. Jie Hu, X. ,. (2014). Two programmed replicative lifespans of Saccharomyces cerevisiae formed by the endogenous molecular-cellular network. Journal of Theoretical Biology .
    8. Maria Stella Carro, W. K. (2010, 1 21). The transcriptional network for mesenchymal transformation of brain tumours. Nature , pp. 463, 318-325 .
    9. Metablic Network. (2014, 6 7). Retrieved 6 7, 2014, from Wikipedia: http://en.wikipedia.org/wiki/Metabolic_networks
    10. Murray, J. D. (2002). Mathematical Biology. Springer.
    11. Parallel computing. (2014, 6 7). Retrieved 6 7, 2014, from Wikipedia.
    12. Sainani, K. (2011-2012). PERSONALIZED CANCER TREATMENT: Seeking Cures Through Computation. Simbios , pp. 5-8.
    13. System Biology. (2014, 6 7). Retrieved 6 7, 2014, from Wikipedia.
    14. Wikipedia. (2014, 6 14). System Biology. Retrieved 6 14, 2014, from Wikipedia: http://en.wikipedia.org/wiki/Systems_biology

    QR CODE