高成炎臺灣大學：資訊工程學研究所黃友正Huang, Yu-ChengYu-ChengHuang2007-11-262018-07-052007-11-262018-07-052004http://ntur.lib.ntu.edu.tw//handle/246246/53616建置高效率大量檢體型快速鑑別檢測類症病原群的完整解決系統，是傳染病臨床檢測與流行病學管控防治的重要目標課題。類屬RNA病毒如重症型急性呼吸症候群(SARS)與輕症型流感(Flu)等類症病原防治實例，顯露檢測誤判與疫情擴大的潛在危機！歸納原因包括遺傳高變異性常導致傳統檢測擴增引子失效之偽陰性檢測結果，類症病原群的檢測基因多重目標複雜度則加強鑑別檢測難度。本項論文針對類症病原群鑑別檢測任務需求的關鍵問題提出解決方案，依據重症與輕症等特定病原選入鑑別檢測用途的多重基因目標，建立特異群探針檢測陣列與最小群引子擴增檢體的演算設計與反應雜合等鑑別檢測技術平台。 本項論文實作PDA-UniQ/MS特異群/最小群-探針引子設計演算法，前篩選階段利用四核苷核化(TNN)索引法快速準確篩選特異區段(uni-S)與多用區段(mu-S)，設計符合生物要件規格與避免立體結構等項篩器模組(filter modules)的特異型檢測探針(uni-probes)與多用型擴增引子(mu-primers)，最佳化階段運用最近鄰(Nearest Neighbor)鏈熔溫度與防制交叉雜合等演算模組最佳化特異群(unique set)無交叉雜合檢測探針，並運用增修型簡要遺傳演算法(MCGA)與鏈熔溫度等化器連結子(Tm-equalizer linker)實作設計最小群(minimum set)等化鏈熔溫度擴增引子。PDA-UniQ/MS演算法實作演算的模擬實驗設計結果顯示，針對擴增目標基因群12669條序列，最小群擴增引子計算3.5時有效減省傳統數量幅度高達68.00%，且序列條數增大則減省幅度更大，演算法效能比較則顯示優於公認最佳的DSH演算法之計算5.5天減省幅度63.98%。 PDA-UniQ/MS實作設計之特異群檢測探針與最小群擴增引子，適用於未經與歷經擴增反應之高敏感度電化學生物感測器檢測技術平台。主要規格特點為：特異群檢測探針50-mer寡核苷酸可分別作為陣列探針(array probe)與檢訊探針(detector probe)，應用於催化訊物沈積型(CARD)訊號擴增的電化學檢測方法。其次，最小群擴增引子應用於檢體核酸聚合鏈反應型(PCR)訊號擴增的電化學檢測方法，與特異群檢測探針陣列進行雜合分析；最小群多用型擴增引子採用12-mer寡核苷酸低複雜度偏短設計，提昇多重基因檢測目標的使用性，再者，鏈熔溫度等化器連結子演算檢測目標基因序列無關的8-mer寡核苷酸，針對個別擴增引子外側客製附加成為20-mer，達成等化熔鏈溫度之最小群擴增引子的規格目標，有效簡化實驗操作與提昇反應效率。 本項論文PDA-UniQ/MS演算法實際驗證的小型生物實驗模式，檢測歸類四屬的九株RNA植物病毒特定代表基因，採用最小群擴增引子進行連續式二階段聚合鏈反應，分別包括初階段低溫觸底法(touch down)與續階段高溫定區法(steady state)的聚合鏈反應。研究檢驗證實僅需設計結果檢省幅度38.80%的11條最小群擴增引子即可完整取代9項檢測基因的9對擴增引子，成功地達到個別擴增檢 測植物病毒的目標基因，降低傳統聚合酶鏈反應的引子製備與實驗成本，增進現行方法檢測技術的品質與效能；未來的驗證生物實驗為整合擴增檢體核酸與雜合陣列技術，進行電化學生物感測器的研究驗證。再者，針對類症病原群的PDA-UniQ/MS整體解決方案，係以SARS類症鑑別檢測設計作為實施範例，考量RNA病毒高度演化變異容易因為序列變異造成檢驗錯誤，藉由高度演化變異仍然存有高度保留區段的生物特性，成功針對SARS類症病原群選入的45個目標序列設計69條最小群擴增引子與特異群檢測探針，助益研究開發達成快速檢測目標。The construction of a highly-efficient and rapid detection of symptom-related pathogens has been an important issue in the prevention and control of epidemics. Actual case studies on RNA virus infections, such as Severe Acute Respiratory Syndrome (SARS), have proven what danger misdiagnosis and widespread infection may lead into. Due to the highly mutational nature of RNA viruses, traditional detections by primers often fail as false negative; the difficulty is increased by the complex target genes of symptom-related pathogens. The PDA-UniQ/MS algorithm uses the “4-mer” hash table by Tetra-Nucleotide Nucleation (TNN) to implement the nearest-neighbor melting temperature calculation and the structure filters. Then, it uses the modified compact genetic algorithm (MCGA) to design the unique probe (uni-probe) and multiple-use primers (mu-primers). The implemented result shows PDA-UniQ/MS greatly reduce the amount of mu-primers needed. In our trial of 12669 amplified target sequences, the reduction rate may reach up to 68% of that of conventional experiments. Further, the result of extended simulation proves that PDA-UniQ/MS will reduce more mu-primers according to the increasing number of target sequences The unique probes and minimum set of mu-primers produced by this PDA-UniQ/MS algorithm is applicable to high-sensitivity electrochemical biosensor platforms either with or without amplification. The benefit of this is that: first, the unique probes, 50-monomer oligonucleotides, can be used as both array probe and detector probe, applying the catalyzed reporter deposition (CARD) electrochemical technique; second, the minimum set of primers, 12-monomer oligonucleotides, applying the polymerase chain reaction (PCR) electrochemical technique, increases the reusability of primers on multiple target genes, which reduces the cost on primer production. Using non-target-related sequences to extra design an 8-monomer oligonucleotide linker for mu-primers, we may minimize the melting temperature range of primers in the same set to simplify the process of the experiment and improve the efficiency of a reaction. Actual detection of plant viruses which are 9 particular representative genes among 4 genuses is done by using the minimum set of mu-primers to implement touch-down PCR on 12-monomer primer fragments and steady state PCR on the total 20-monomer mu-primer with linker to amplify the samples. The experimental result proves the 11 mu-primers can successfully amplify the 9 target genes of plant viruses instead of traditional 18 primers. It shows PDA-UniQ/MS can greatly reduce the cost of PCR experiment and improve both the quality and efficiency of the presenting technology. In the future, a hybridization experiment integrated with the labeled nucleotides of pathogen is the validation of electrochemistry biosensor. Concerning the design of symptom-related pathogens, we noted that, as with the case of SARS virus, the high mutation rate of RNA viruses must be considered, or the detection would likely fail due to sequence mutation. Based on the biological characteristic of highly-conserved regions, PDA-UniQ/MS algorithm successfully designed 73 mu-primers to amplify 50 target sequences. It also produced unique detection probes according to biological structure characteristics and probe specificity, so that rapid detection may be achieved.Chapter 1 Introduction 1 Chapter 2 UniQue Probe and Multiple-use Primer Design Algorithm 3 2.1 Introduction 3 2.2 Overview 4 2.3 Uni-Probe and Mu-primer Design Algorithm at Target Sections 6 2.3.1 Unique section and Multiple-use section (US&MS) finding and Tm calculation by Nearest-Neighbor Model 6 2.3.2 Uni-probe design at US 8 2.3.3 Mu-primer design at MS 9 2.3.4 Design Linker for Mu-primer 10 2.3.5 Mu-primers Design for Touch-down PCR Process 10 2.4 Case Study: Plant Virus Model Experiment 11 2.4.1 Design uni-probe and mu-primer for plant viruses by PDA-UniQ/MS 11 2.4.2 Linker design for Fasten PCR Process 13 2.4.3 Mu-primer result (1-D gel) 14 2.4.4 The extended simulation of plant virus by PDA-UniQ/MS 15 2.4.5 The extended simulation of linker design for Mu-primer of plant virus 16 2.5 Discussion 21 Chapter 3 Design uni-probes at US and mu-primers at MS 22 3.1 Introduction 22 3.2 DNA encoding 23 3.3 Selection criteria 24 3.4 Search unique segments 25 3.5 Find unique sections 26 3.6 Nearest-Neighbor model 27 3.6.1 Specificity of an oligonucleotide candidate 27 3.6.2 Nearest-neighbor model 28 3.6.3 Melting Temperature Calculation 29 3.6.4 Implementation 29 3.7 Probe and Primer Design from TNN Indexing 32 3.8 Simulation Result 33 Chapter 4 MCGA for Mu-primer design 36 4.1 Introduction 36 4.2 Multiple-Use PCR Primer Design Problems 38 4.2.1 Problem Definition 38 4.2.2 Problem Transformation 39 4.3 Modified Compact Genetic Algorithm (MCGA) 41 4.3.1 Chromosome representation 41 4.3.2 The local search 41 4.3.3 The competition mechanism 42 4.4 Experimental Result 42 4.4.1 The result of set-covering problem 43 4.4.2 The improvement in reducing needed primers of LTH within acceptable calculation time 43 4.5 Conclusion and Future Work 47 Chapter 5 Conserved region for RNA virus - Applied Differential Detections on Multigene Targets of Symptom-related RNA Viruses Group 49 5.1 Introduction 49 5.2 Finding conserved region from multiple sequence alignment 54 5.3 Simulation Result 55 Chapter 6 Conclusions 56 6.1 Summary 56 6.2 Future work 57 Reference . 59 List of Publication 62 Appendix (I) All SARS sequence 63 Appendix (II) Corona Virus multiple sequence alignment (MSA) result 66 Appendix (III) Symptom-based Etiology Group design 845406434 bytesapplication/pdfen-US多重目標基因早期鑑別檢測類症病原群快速早期鑑別檢測最小集合解之演算法聚合脢鍊反應引子設計探針設計生物晶片遺傳演算法流行感冒病毒早期鑑別檢測Probes designGenetic AlgorithmMinimum Set Covering ProblemPCR Primer designEarly virus detectionBiosensor designMicroarray probe designMultigene Targets of Symptom-Related Pathogens detEarly Differential Detectionsthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/53616/1/ntu-93-R91922021-1.pdf