指導教授：邱繼輝臺灣大學：生化科學研究所王博偉Wang, Po-WeiPo-WeiWang2014-11-262018-07-062014-11-262018-07-062014http://ntur.lib.ntu.edu.tw//handle/246246/261653質譜儀是一種用來鑑定未知化學分子，包括醣質結構的一個非常有用的分析工具。目前的醣質體分析已經漸漸地從單面相基質輔助雷射質譜(MALDI MS)分析轉向連續性的液相層析電噴灑離子化(LC-ESI-MS/MS)串聯質譜分析，以便可以針對偵測到的醣盡可能地做更多MS2及MS3，來處理及更深度分析較複雜的樣品。一般來說，從LC-MSn得到的數據會比從MALDI MS得到的大很多。這使得以人工分析質譜數據及標註圖譜非常耗時且乏味，而且對要進入此領域的非專家而言有相當高的技術門檻。因此，高效率且能自動化處理LC-MSn醣質體分析數據的軟體為目前的急需。 本論文的研究主要在開發一個可以用來全面性、並且高效率地找出醣斷片/醣末端表位及它們的組成的軟體。為能有效地接受並整合所用的質譜儀產生的數據格式，我們主要使用微軟的C#.net 架構來開發此軟體，將其命名為GlyPick，並應用在處理來自於胃癌細胞甲基化醣質的LC-MSn資料 。GlyPick能辨認並找出所有MS2圖譜裡面的斷片/末端表位，將其對應到MS1的母離子，修正該母離子的正確monoisotopic m/z 質，以便算出該質量所對應的醣分子; 另外，也將該MS2裡面的斷片/末端表位對應到其MS3的圖譜，得以鑑定有關連接及結構的資訊。最後，再整合所有的資料取得最符合MS1的m/z以及MS2/MS3的斷片/末端表位的醣結構。GlyPick也將所有鑑定的醣質組合根據液相層析時間來分組歸類，以便更方便地看出結果，更深入地分析數據。將GlyPick應用於胃癌細胞的N連接的、O連接的以及連接於脂肪的醣，可以顯示出醣斷片/末端表位以及其醣質在這些樣品中的差別，並以此應用來評估 GlyPick和幾個目前已經發表的醣質譜數據分析工具的功能及效率上的差異。Mass spectrometry (MS) is one of the most useful tools for analyzing unknown chemical molecules and structures including glycans. Current glycomic analysis has gradually moved from matrix-assisted laser desorption/ionization (MALDI)-based offline analysis to online liquid chromatography (LC)-electrospray ionization (ESI)-based MS/MS applications, in order to cope with the sample complexity and the need to do as many MS2 and MS3 as possible for each of the detected glycan precursors. Generally speaking, the datasets acquired by LC-MSn are much larger than the ones of MALDI-MS. This makes manual identification and annotation of glycans quite tedious, time-consuming, and not amenable to non-experts. Therefore, efficient softwares are urgently needed to automate part or all of the data analysis process in LC-MSn-based glycomics. In this thesis work, a software for comprehensive, efficient glycosyl epitope/fragment ions discovery and glycosyl composition annotation was developed using C# .net framework for better integration with Thermo Raw API, and the easy-to-maintain object-oriented concept. Applied to LC-MSn datasets acquired on the permethylated glycans derived from a gastric adenocarcinoma cell line AGS, we demonstrated the utilities of this computational tool, named GlyPick. GlyPick can recognize and discover specific fragment ions or epitopes in MS2 spectra, tracking back to MS1 spectra to locate the monoisotopic precursor, assign its glycosyl composition so as to confirm the fragment ions/epitopes indeed derived from glycans, and then tracking forward to MS3 spectra for more information on linkages. By integrating the data, the most plausible glycan structures consistent with the deduced monoisotopic precursor peaks will be assigned in accordance with the MS2-detected epitopes/fragment ions, complete with linkage determination in cases where MS3 were productively acquired. GlyPick also groups the fitted glycosyl compositions by retention times for better visualization and further data mining. Applications of GlyPick to datasets of N-glycans, O-glycan and glycolipids from AGS indicated differences of epitopes/fragment ions and glycosyl compositions in these samples. Finally, we also compared GlyPick with other published tools, MultiGlycan, GlycReSoft and SimGlycan, to evaluate their performance differences.口試委員審定書 I 致謝 II 摘要 III ABSTRACT IV LIST OF FIGURES VIII LIST OF TABLES XIII 1.INTRODUCTION 1 1.1 Biological importance of glycosylation 1 1.2 The diversity and complexity of glycans 3 1.3 Mass spectrometry-based glycomics 6 1.4 Current glycan analytical tools and limitations 17 1.5 The aims of GlyPick 21 2. MATERIALS AND METHODS 23 2.1 Glycan standards, Sample sources and preparation from tissues or cell lines and Mass Spectrometry analysis 23 2.2 Softwares development tools and running environment 24 3. RESULT 27 3.1 The rationale underlying GlyPick development 27 3.2 The overall workflow of GlycPick 32 3.3 The processing unit and algorithms of software 33 3.4 Modules of software 44 3.5 The GUI(Graphical User Interface) and details of comma-separated values (CSV) files of GlyPick 53 3.6 Evaluation of software 56 4. DISCUSSION 83 4.1 What GlyPick can and can not do? 83 4.2 Comparisons between glycomics softwares 84 4.3 Future perspectives for MS2-Based spectrum analytical softwares 85 4.4 Future perspectives of glycomic informatics 86 REFERENCES: 88 ABREVIATIONS: 9113198235 bytesapplication/pdf論文公開時間：2019/03/09論文使用權限：同意有償授權(權利金給回饋學校)質譜儀醣質結構醣質譜數據分析工具醣斷片/末端表位液相層析電噴灑離子化(LC-ESI-MS/MS)串聯質譜分析胃癌細胞甲基化醣質thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/261653/1/ntu-103-R00b46014-1.pdf