臺灣博碩士論文加值系統

胞吐作用為蛋白質運輸、神經傳導和激素分泌的重要途徑。藉由先進的顯微和螢光染色的技術，生物學家能於更微小的時間、空間解析度下觀測單一胞吐作用中運輸、定位(docking)、預備(priming)、融合及回收等過程。由於每個細胞中有數百到數千個囊泡，如果沒有輔助工具，生物學家很困難同時對囊泡的特性進行定性判定與量化。
目前文獻中已知的系統僅能提供10-25%囊泡的資訊，導致生物學家無法獲得完整資訊對研究結做出正確的判讀。本研究透過改善影像處理及軌跡追蹤等方法，建構高準確度的自動化螢光囊泡蛋白影像之追蹤輔助系統，全程紀錄囊泡蛋白質的亮度、面積及軌跡變化，使生物學家對其生理機制有完整而正確的判定。針對PC12及βTC3細胞影像中囊泡大小與背景不一，本研究以不同濾波器處理影像雜訊，再採用雙閥值形態學濾波法使系統能正確框選細胞影像大部份大小囊泡區塊。對於移動速度不同的囊泡，本研究中的追蹤法以最短距離連接軌跡或以Kalman filter 輔助預測囊泡位置，最後以階層對應法找到最可能連接之囊泡移動路徑。
由研究結果顯示，以最短距離為追蹤方法的PTrack系統追蹤移動情況少的PC12細胞影像，其正確率比PTrack II多20%。囊泡移動模擬結果中，加入Kalman filter之PTrack II比PTrack增加9~28%追蹤正確率。以βTC3細胞影像驗證追蹤效能，PTrack II之正確率可達56%，比PTrack增加26%。經詳細分析結果後，增加的正確率中大多是速度快的大囊泡，證明PTrack II可以追蹤大面積且移動快速的囊泡。本研究建構的追蹤系統，可以大幅度地提升對囊泡蛋白在活細胞中的動態特質的定性與定量的分析，提供生物學家更完整的資訊。

Membrane trafficking is a very important physiological process involved in protein transport, endocytosis, and exocytosis. The functions of vesicles are strongly correlated with various spatial dynamic properties of vesicles, including their types of movements and morphology. There are hundreds to thousands of synaptic vesicles near the plasma membrane that biologist is diffuclt to capture all the information without any auxiliary tools. The purpose of the research is to help biologist to analyze the dynamic property of the vesicles. This study builds a highly accuracy automated vesicle tracking system by improving the image process and tracking algorighm that records the complete information of vesicle with their physiological mechanism. To identify the vesicle's feature in cell image, the system used different image filters to deal with image noise first, the morphological filter with two-threshold image processing techniques was used to locate granules of various vesicle size in PC12 and βTC3 cells. For those fast-moving and large granules, Kalman filtering was used to improve the performance in tracking process. Performance was evaluated by using five simulations and two cells images (PC12 and βTC3). PTrack got 20% accuracy more than PTrack II after evaluation in PC12 cells images, and the exocytosis event was captured by PTrack exactly. PTrack II was validated using time-lapse images of insulin granules in βTC3 cells, which revealed that PTrack II could track better than PTrack, averaged accuracy up to 56%. The overall tracking results indicate that PTrack get better tracking performance in immobile vesicles, PTrack II is better at tracking vesicles with various dynamic properties, which will facilitate the acquisition of more-complete information on vesicle dynamics.

目錄
摘要 I
ABSTRACT II
誌謝 III
目錄 IV
圖目錄 VIII
表目錄 XI
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機 2
1.2.1 蛋白質定位資訊的重要 2
1.2.2 蛋白質定位與疾病 3
1.2.3 胞吐作用(Exocytosis)含有大量複雜的蛋白質動態定位資訊 3
1.2.4 囊泡蛋白質之動態定位分析的困難 5
1.3 研究目的 5
第二章 蛋白質研究與文獻回顧 7
2.1 蛋白質分析研究方法 7
2.2 手動囊泡蛋白質追蹤系統 - METAMORPH 8
2.3 相關自動囊泡追蹤研究方法 9
2.3.1 影像特徵擷取 9
2.3.2 追蹤法則 9
2.3.3 Kalman filter 10
2.3.4 資料關聯(Data association) 11
2.4 自動分析追蹤蛋白質相關研究之比較 12
2.5 文獻回顧總結 13
第三章 理論基礎 15
3.1 囊泡蛋白質影像擷取 15
3.1.1 雷射共軛焦(Confocal)顯微鏡 18
3.1.2 全反射螢光顯微鏡(TIRFM) 18
3.2 數位影像處理 20
3.2.1 影像前處理濾波器選用 20
3.2.2 Morphological filter 22
3.2.3 二值化影像 25
3.3 軌跡追蹤 26
3.3.1 追蹤對應 26
3.3.2 Kalman filter 28
3.4 蛋白質特性分析 34
3.4.1 亮度與面積-定量分析 34
3.4.2 Mean Square Displacement (MSD)-定性分析 34
第四章 系統架構與研究方法 37
4.1 蛋白質影像 37
4.2 系統架構 38
4.3 PTRACK 39
4.4 PTRACK II 40
4.4.1 影像處理及辨識 40
4.4.2 PTrack II軌跡追蹤 41
4.4.3 初始軌跡建立 42
4.4.4 搜尋區域 42
4.4.5 細胞軌跡對應 43
4.5 簡易結果分類與判斷 44
4.6 實驗驗證與設計 46
4.6.1 模擬驗證設計 46
4.6.2 真實影像追蹤驗證設計 47
第五章 結果與討論 49
5.1 模擬測試結果 49
5.1.1 Simulation 1 - 胞吐影像模擬 49
5.1.2 Simulation 2 - MSD模擬測試 50
5.1.3 Simulation 3 -不同MSD型態在不同速度下之模擬 52
5.1.4 Simulation 4 -兩囊泡交錯軌跡測試 53
5.1.5 Simulation 5 -囊泡密度測試 56
5.2 PC12細胞胞吐影像追蹤 58
5.2.1 PC12細胞囊泡胞吐亮度變化 59
5.2.2 PC12細胞影像追蹤結果比較 62
5.3 ΒTC3細胞中INSULIN囊泡追蹤驗證 65
5.3.1 βTC3細胞影像辨識 65
5.3.2 βTC3細胞影像追蹤結果 66
5.3.3 囊泡麵積vs囊泡移動速度結果分析 68
5.3.4 囊泡運動形態結果分析 69
5.4 討論 70
第六章 結論與未來展望 73
6.1 結論 73
6.2 未來展望 74
參考文獻 76


圖目錄
圖 1 1 蛋白質行為分析 4
圖 2 1不同蛋白質分析方法特性 7
圖 3 1 ECLIPTIC GFP 被488NM激發之光譜 16
圖 3 2 ECLIPTIC PHLUORIN 被融化到N-TERMINUS 17
圖 3 3 螢光顯微鏡基本原理架構圖 17
圖 3 4 反射螢光影像 18
圖 3 5 顯微影像(A)全反射螢光影像(TIRFM) (B)一般顯微鏡影像 19
圖 3 6 EGFP SYNAPTICPHLUORINS 胞吐影像序列. 20
圖 3 7 以WIENER FILTER 濾除雜訊(A)原圖(B)濾除雜訊後結果 22
圖 3 8 膨脹運算過程 23
圖 3 9 侵蝕運算過程 23
圖 3 10 TOP HAT TRANSFORMATION 25
圖 3 11 TOP HAT範例圖(A)原圖 (B) OPENED IMAGE (C)結果 25
圖 3 12 255階的影像影像G1 26
圖 3 13 二值化後的G2影像 26
圖 3 14 離散系統方塊圖 30
圖 3 15 KALMAN FILTER循環圖 32
圖 3 16 離散KALMAN FILTER方塊圖 33
圖 3 17 KALMAN FILTER運作圖 33
圖 3 18 蛋白質影像亮度與面積變化 34
圖 3 19 MSD回歸分析圖 35
圖 4 1 NPY-EGFP囊泡影像 37
圖 4 2 蛋白質囊泡亮度變化 38
圖 4 3 ΒTC3細胞的INSULIN GRANULES 38
圖 4 4 系統方塊圖 39
圖 4 5 PTRACK 系統流程 40
圖 4 6 PTRACK II 影像處理流程 41
圖 4 7 PTRACK II 影像處理示範圖 41
圖 4 8 PTRACK II 軌跡追蹤流程圖 41
圖 4 9 運用階層對應法與KALMAN FILTER追蹤 44
圖 4 10 錯誤追蹤範例 44
圖 4 11 簡易追蹤結果分類 45
圖 5 1 胞吐影像模擬測試 49
圖 5 2 RANDOM DIFFUSION類型模擬 50
圖 5 3 DIRECTED DIFFUSION類型模擬 51
圖 5 4 CAGED DIFFUSION類型模擬 51
圖 5 5 MOTION 1 53
圖 5 6 MOTION 2 54
圖 5 7 MOTION 3 54
圖 5 8 MOTION 4 55
圖 5 9 囊泡密度追蹤測試結果 57
圖 5 10 囊泡亮度變化分析 60
圖 5 11 NPY-EGFP 囊泡分泌追蹤範例 61
圖 5 12 EGFP-RAB3A 62
圖 5 13 NPY-EGFP 囊泡追蹤 63
圖 5 14 SYNAPTOPHLUORIN 囊泡追蹤 64
圖 5 15 PEROXISOMES 囊泡追蹤 65
圖 5 16 不同閥值取得特徵之結果 66
圖 5 17 囊泡麵積及移動速度分佈圖 69
圖 5 18 囊泡運動形態分析 70
圖 5 19 運動形態分析結果不同探討 70
圖 5 20 以WATH WINDOW監測消失的囊泡 71
圖 5 21 PTRACK II追蹤軌跡 72


表目錄
表 1 1 蛋白質化學、蛋白質體學和蛋白質動態分析比較 1
表 2 1 相關研究使用之影像處理及追蹤方法 13
表 3 1 用來標定細胞器官或胞吐作用的GFP (或其變體) 15
表 4 1 追蹤結果分類表 45
表 4 2 追蹤結果分類說明 45
表 5 1 CAGED DIFFUSION 52
表 5 2 DIRECTED DIFFUSION 52
表 5 3 RANDOM DIFFUSION 52
表 5 4 DIRECTED及CAGED DIFFUSION交互測試表 53
表 5 5 MOTION 1測試結果 55
表 5 6 MOTION 2測試結果 56
表 5 7 MOTION 3測試結果 56
表 5 8 MOTION 4測試結果 56
表 5 9 密度測試速度變異量 58
表 5 10 四組細胞胞吐影像第一張圖所辨識到之囊泡數及追蹤正確數 58
表 5 11 四組細胞 LINE-LINKING 正確率(%) 59
表 5 12 四組細胞 FRAME-LINKING 正確率(%) 59
表 5 13 METAMORPH 與 PTRACK比較 62
表 5 14 六組細胞影像第一張圖所辨識到之囊泡數及追蹤正確數 67
表 5 15 ΒTC3 LINE-LINKING 正確率(%) 67
表 5 16 ΒTC3 FRAME-LINKING 正確率(%) 67

[1]Daniel CL. Introduction to Proteomics-Tools for the New Bioloogy: Humana Press Inc; 2002.
[2]Nobel Foundation Press Release. 1999 [cited; Available from: http://nobelprize.org/nobel_prizes/medicine/laureates/1999/
[3]Mochizuki N, Yamashita S, Kurokawa K, Ohba Y, Nagai T, Miyawaki A, et al. Spatio-temporal images of growth-factor-induced activation of Ras and Rap1. Nature 2001;411:1065-8.
[4]Heo WD, Meyer T. Switch-of-function mutants based on morphology classification of Ras superfamily small GTPases Cell. Cell 2003;113:315-28.
[5]Eils R, Athale C. Computational imaging in cell biology. J Cell Biol 2003;161:477-81.
[6]Murthy VN, Stevens CF. Reversal of synaptic vesicle docking at central synapses. NatNeurosci 1999;2(6):503-7.
[7]Toomre D, Steyer JA, Keller P, Almers W, Simons K. Fusion of constitutive membrane traffic with the cell surface observed by evanescent wave microscopy. J Cell Biol 2000 Apr 3;149(1):33-40.
[8]Taraska JW, Perrais D, Ohara-Imaizumi M, Nagamatsu S, Almers W. Secretory granules are recaptured largely intact after stimulated exocytosis in cultured endocrine cells. ProcNatlAcadSciUSA 2003;100:2070-5.
[9]Rizzoli SO, Betz WJ. Synaptic vesicle pools. NatRevNeurosci 2005;6(1):57-69.
[10]Albeck JG, MacBeath G, White FM, Sorger PK, Lauffenburger DA, Gaudet S. Collecting and organizing systematic sets of protein data. Nat Rev Mol Cell Biol 2006 Nov;7(11):803-12.
[11]Boland MV, K. MM, Murphy RF. Automated recognition of patterns characteristic of subcellular structures in fluorescence microscopy images. Cytometry 1998;33:366-75.
[12]Tseng Y, Wirtz D. Mechanics and multiple-particle tracking microhetrogeneity of alpha-actinin-cross-linked actin filament networks. BiophysJ 2001;81:1643-56.
[13]Porntula P, Denis W, Yiider T. Structure-function relationship of biological gels revealed by multiple-particle tracking and differential interference contrast microscopy: The case of human lamin networks. The American Physical Society 2004;70:041906-1--7.
[14]Mica O-I, Chiyono N, Toshiteru K, Shintaro N, Yoko N, Shinya N. TIRF imaging of docking and fusion of single insulin granule motion in primary rat pancreatic ?-cells: different behaviour of granule motion between normal and Goto-Kakizaki diabetic rat ?-cells. BiochemJ 2004;381:13-8.
[15]J.Suh, D.Wirtz, J.Hanes. Real-time intracellular transport of gene nanocarriers studied by multiple particle tracking. BiotechnolProg 2004;20:598-602.
[16]Junghae S, Michelle D, Justin H. Real-time multiple-particle tracking: applications to drug and gene delivery. Advanced Drug Delivery Reviews 2005;57:63-78.
[17]M.Dawson, E.Krauland, D.Wirtz JH. Transport of polymeric nanoparticle gene carriers in gastric mucus. BiotechnolProg 2004;20:851-7.
[18]Li X, Hao M, Piston DW, Dawant BM. Automatic tracking of proteins in sequences of fluorescence images. P Soc Photo-Opt Inst 2004;5370:1364-71.
[19]Cheezum MK, Walker WF, Guilford WH. Quantitative comparison of algorithms for tracking single fluorescent particles. Biophys J 2001 Oct;81(4):2378-88.
[20]Gelles J, Schnapp, B.J. Tracking kinesin-driven movements with nanometer scale precision. Nature 1988;331:450-3.
[21]Carter BC, Shubeita GT, Gross SP. Tracking single particles: a user-friendly quantitative evaluation. Phys Biol 2005 Mar;2(1):60-72.
[22]黃芸妮. 螢光蛋白質自動辨識及追蹤系統. 中原大學生物醫學工程研究所碩士輪文 2004.
[23]胡志承. 利用機器視覺於水產精子之活動性分析. 臺灣大學生物產業機電工程學研究所碩士輪文 2003.
[24]Apgar J, Tseng Y, Fedorov E, Herwig MB, Almo SC, Wirtz D. Multiple-particle tracking measurements of heterogeneities in solutions of actin filaments and actin bundles. Biophys J 2000 Aug;79(2):1095-106.
[25]Genovesio A, Liedl T, Emiliani V, Parak WJ, Coppey-Moisan M, Olivo-Marin JC. Multiple particle tracking in 3-D+t microscopy: method and application to the tracking of endocytosed quantum dots. IEEE Trans Image Process 2006 May;15(5):1062-70.
[26]Crocker JC, Grier DG. Methods of Digital Video Microscopy for Colloidal Studies J Colloid Interf Sci 1996;179:298-310.
[27]Mohinder S. Grewal APA. Kalman Filtering : Theory and Practice Using MATLAB 2 edition: Wiley-Interscience; 2001.
[28]Kalman RE. A new approach to linear filtering and prediction problems. J Basic Eng-T ASME 1960;82(Series D):35-45.
[29]Beresford-Smith B, Helden DFV. Applications of radar tracking algorithms motion analysis in biomedical images. Image Processing, Proceedings; 1994 1994; 1994. p. 411-5.
[30]Cox IJ, Hingorani SL. An efficient implementation of Reidapos;s multiple hypothesis trackingalgorithm and its evaluation for the purpose of visual tracking. IEEE Transactions Pattern Analysis and Machine Intelligence, 1996;18(2):138 - 50.
[31]Bar-Shalom Y, Blair WD. Multitarget-Multisensor Tracking Applications and Advances: Norwood, MA: Artech House; 2000.
[32]Tvarusko W, Bentele M, Misteli T, Rudolf R, Kaether C, Spector DL, et al. Time-resolved analysis and visualization of dynamic processes in living cells. ProcNatlAcadSciUSA 1999;96:7950-5.
[33]Vallotton P, Ponti A, Waterman-Storer CM, Salmon ED, Danuser G. Recovery, visualization, and analysis of actin and tubulin polymer flow in live cells: a fluorescent speckle microscopy study. Biophys J 2003;85:1289-306.
[34]Hao M, Li X, Rizzo MA, Rocheleau JV, Dawant BM, Piston DW. Regulation of two insulin granule populations within the reserve pool by distinct calcium sources. J Cell Sci 2005;118:5873-84.
[35]Nofal S, Becherer U, Hof D, Matti U, Rettig J. Primed vesicles can be distinguished from docked vesicles by analyzing their mobility. J Neurosci 2007;27(6):1386-95.
[36]Chalfie M, Kain S. Green Fluorescent Protein Properties, Applications, and Protocols. Cananda: Wiley-liss; 1998.
[37]Ormo M, Gobitt AB, kallio k, Gross LA, Tsien RY, Remington SJ. Grystal structure of Aequorea Victoria green fluorescent protein. Science 1994;273:1392-5.
[38]Rost FWD. Fluorescent Microscopy: Cambridge University Pross; 1992.
[39]Ponti A, Vallotton P, Salmon WC, Waterman-Storer CM, Danuser G. Computational analysis of F-actin turnover in cortical actin meshworks using fluorescent speckle microscopy. Biophys J 2003 May;84(5):3336-52.
[40]Serra J. Image analysis and mathematical morphology. London: Academic Press,London; 1982.
[41]Serra J. Image analysis and mathematical morphology:theoretical advances. London: Academic Press; 1998.
[42]Sonka M, Hlavac V, Boyle R. Image Processing, Analysis, and Machine Vision: Cole Publishing Company 1999.
[43]Rangarajan K, Shah M. Establishing motion correspondence. Computer Vision and Pattern Recognition; 1991 1991/06/06/; 1991. p. 103-8.
[44]Sethi JK, Jain R. Finding trajectories of feature points in a monocular image sequence. IEEE T Pattern Anal 1987;9:56-73.
[45]Veenman CJ, J.T.Reinders, E.Backer. Resolving motion correspondence for densely moving point. IEEE Transaction on Pattern Analysis and Machine Intelligence 2001;23(1):54-72.
[46]Kevin M. Kalman filter toolbox for Matlab. 2004 [cited; Available from:
[47]Qian H, Sheetz MP, Elson EL. Single particle tracking. Analysis of diffusion and flow in two-dimensional systems. Biophys J 1991;60(4):910-21.
[48]Li D, Xiong J, Qu A, Xu T. Three-dimensional tracking of single secretory granules in live PC12 cells. Biophys J 2004;87:1991-2001.
[49]Saxton MJ, Jacobson K. Single-particle tracking: applications to membrane dynamics. Annu Rev Bioph Biom 1997;26:373-99.
[50]Ku TC, Huang YN, Huang CC, Yang DM, Kao LS, Chiu TY, et al. An automated tracking system to measure the dynamic properties of vesicles in living cells. Microsc Res Techniq 2007;70(2):119-34.
[51]Chetverikov D, Verestoy J. Feature point tracking for incomplete trajectories. Computing 1999;62(4):321-38.
[52]Becherer U, Moser T, Stuhmer W, Oheim M. Calcium regulates exocytosis at the level of single vesicles. Nat Neurosci 2003 Aug;6(8):846-53.