Dissertations / Theses on the topic 'Sequential and selective labeling'

The graph isomorphism problem has received a great deal of attention on both theoretical and practical fronts. However, a polynomial algorithm for the problem has yet to be found. Even so, the best of the existing algorithms perform well in practice; so well that it is challenging to find hard instances for them. The most efficient algorithms, for determining if a pair of graphs are isomorphic, are based on the individualization-refinement paradigm, pioneered by Brendan McKay in 1981 with his algorithm nauty. Nauty and various improved descendants of nauty, such as bliss and saucy, solve the graph isomorphism problem by determining a canonical representative for each of the graphs. The graphs are isomorphic if and only if their canonical representatives are identical. These algorithms also detect the symmetries in a graph which are used to speed up the search for the canonical representative--an approach that performs well in practice. Yet, several families of graphs have been shown to exist which are hard for nauty-like algorithms. This dissertation investigates why these graph families pose difficulty for individualization-refinement algorithms and proposes several techniques for circumventing these limitations. The first technique we propose addresses a fundamental problem pointed out by Miyazaki in 1993. He constructed a family of colored graphs which require exponential time for nauty (and nauty's improved descendants). We analyze Miyazaki's construction to determine the source of difficulty and identify a solution. We modify the base individualization-refinement algorithm by exploiting the symmetries discovered in a graph to guide the search for its canonical representative. This is accomplished with the help of a novel data structure called a guide tree. As a consequence, colored Miyazaki graphs are processed in polynomial time--thus obviating the only known exponential upper-bound on individualization-refinement algorithms (which has stood for the last 16 years). The preceding technique can only help if a graph has enough symmetry to exploit. It cannot be used for another family of hard graphs that have a high degree of regularity, but possess few actual symmetries. To handle these instances, we introduce an adaptive refinement method which utilizes the guide-tree data structure of the preceding technique to use a stronger vertex-invariant, but only when needed. We show that adaptive refinement is very effective, and it can result in dramatic speedups. We then present a third technique ideally suited for large graphs with a preponderance of sparse symmetries. A method was devised by Darga et al. for dealing with these large and highly symmetric graphs, which can reduce runtime by an order of magnitude. We explain the method and show how to incorporate it into our algorithm. Finally, we develop and implement a parallel algorithm for detecting the symmetries in, and finding a canonical representative of a graph. Our novel parallel algorithm divides the search for the symmetries and canonical representative among each processor, allowing for a high degree of scalability. The parallel algorithm is benchmarked on the hardest problem instances, and shown to be effective in subdividing the search space.
Ph.D.
School of Electrical Engineering and Computer Science
Engineering and Computer Science
Computer Science PhD

Arterial Spin Labeling (ASL) is a Magnetic Resonance Imaging (MRI) technique that is able to non-invasively quantify the rate of delivery of arterial blood to tissue, known as perfusion. In this thesis a method that uses Velocity Selective (VS) preparations to generate contrast between blood and tissue spins is investigated. The systematic errors associated with performing a VSASL experiment on imperfect hardware is first investigated. It is shown through simulations and experiments that some VS preparations will underestimate perfusion due to static and transmit magnetic field errors, and that eddy currents caused by switching of magnetic gradients lead to an overestimation of perfusion with VSASL by up to a factor 2. A novel VS preparation, BIR-8, is presented which is shown to be the most robust to these imperfections. The BIR-8 VSASL technique is then applied in brain tumours where it is found that significant VSASL signal can be detected in less than 5 minutes. However, in a comparison with a spatially selective ASL technique it is found that VSASL overestimates perfusion in these tumours, despite agreeing in Grey Matter. The systematic errors due to physiology are then modelled, and it is shown that both diffusion and bulk motion will systematically bias the VSASL measurement. A diffusion insensitive VSASL technique, VS-TILT, is then developed and it is found that a significant proportion of the VSASL signal originates from diffusion effects. Theoretical models for the shape of the bolus in vascular networks are also derived, and it is shown that an isotropic network of laminar vessels produces the most efficient saturation, but saturation is also achieved with plug flow. The diffusion insensitive VS preparation is then applied in an attempt to isolate the venous compartment in order to measure Oxygen Extraction Fraction. A kinetic model is derived in order to optimise the acquisition. However, it is found that accurate measurements of OEF would not be produced by this sequence in a clinically realistic time.

This dissertation develops, applies, and investigates new methods to improve the analysis of logistic regression mixture models. An interesting dose-response experiment was previously carried out on a mixed population, in which the class membership of only a subset of subjects (survivors) were subsequently labeled. In early analyses of the dataset, challenges with separation of points and asymmetric confidence intervals were encountered. This dissertation extends the previous analyses by characterizing the model in terms of a mixture of penalized (Firth) logistic regressions and developing methods for constructing profile likelihood-based confidence and inverse intervals, and confidence bands in the context of such a model. The proposed methods are applied to the motivating dataset and another related dataset, resulting in improved inference on model parameters. Additionally, a simulation experiment is carried out to further illustrate the benefits of the proposed methods and to begin to explore better designs for future studies. The penalized model is shown to be less biased than the traditional model and profile likelihood-based intervals are shown to have better coverage probability than Wald-type intervals. Some limitations, extensions, and alternatives to the proposed methods are discussed.

Microbubble mediated technologies are employed for pretreatment steps due to the characteristics of the gas-liquid interface. Traditionally, pretreatment processes are energy intensive operations and use hazardous chemicals such as sulfuric acid and hydrochloric acid, which need to be removed from the pretreatment slurry before feeding it to the fermentation process. Alternative approaches to microbubbles for pretreatment, however, have significant challenges. For example, conventional bubbles are several orders of magnitude larger than the bubble exit pore and therefore have less direct contact with the biomass or delivery the ozone efficiently to the pretreatment slurry. Consequently, these concerns have been addressed in this research, and microbubble-microbe synergy and Ozonolysis-microbe synergy for biomass pretreatment with the developing of microbubbles driven systems, were used to facilitate microbubble generation suitable for pretreatment processes. The first approach was achieved by exploiting the synergy between microbubble-microbe to pretreat lignocellulosic biomass and glucose was the target product. The effects of microbubbles, microbe and the synergy between them on morphology, functional groups and glucose yield were investigated. It was found that free radicals around the gas-liquid interface of the microbubble can readily attack and degrade lignocellulosic biomass, rendering it more amenable to digestion. The combination of microbubbles and Pseudomonas putida—a robust delignification and cellulolytic microbe, further improved biomass degradation and consequently, increased glucose production from wheat straw in comparison to solo pretreatment of the biomass with microbubbles and Pseudomonas putida respectively. In addition, it was found that the highest glucose achieved was 0.27 mg/ml. The second was conducted by exploiting ozonolysis-microbe synergy to pretreat lignocellulosic biomass and glucose was also the target product. The effects of ozonation at various pHs and ozone concentrations, biological pretreatment by Pseudomonas putida and the synergy between them on morphology, functional groups and glucose yield were explored. Ozone is a strong oxidative agent that reacts with lignin by attacking the carbon-carbon double bonds, while P. putida preferentially hydrolyses the exposed cellulolytic parts of the biomass to simple sugars. It was found that both lignin and cellulose contents were reduced under this pretreatment with relatively high glucose recovery. The highest glucose concentration reached was 1.1 mg/ml after 24 hr ozonation at 8.86 mg/L ozone and pH 3 with 50 % reduction in the biological pretreatment duration but crucially, increasing microbial biomass. Using the synergetic approach for the biomass pretreatment is promising approach but leaves the pretreatment slurry contaminated with the cellulolytic microbe, Pseudomonas putida, which needs to be inactivated or removed before feeding the pretreatment slurry into the fermenter. The ability of carbon dioxide enriched microbubbles to inactivate Pseudomonas putida was subsequently investigated. Many drawbacks of the traditional sterilization methods were avoided by using carbon dioxide enriched microbubbles, such as high energy consumption and using toxic and corrosive reagents. It was found that 2-Log reduction in the bacterial population after 90 min was achieved using carbon dioxide enriched microbubbles. Further reductions were achieved by adding additives such as ethanol and acetic acid and the highest reduction performed was 3.5 Log with 10 % ethanol, while a 2.5-Log reduction was achieved with 0.5 % acetic acid. These reductions in the bacterial population were concurrent with changing cells shape from rod cells to coccus shape with cell damage such as lesions and cells death. Subsequently, aerobic fermentation with glucose as a carbon source proceeded with Zymomonas mobilis ZM4 as the microbial fermentation agent. Acetaldehyde has drawn the attention in this research because it is an important chemical, and it can be used in many processes such as plastic manufacturing and fuels production such as ethanol and butanol. Several attempts to produce acetaldehyde from Zymomonas mobilis or genetically modified microbes contained some genes from Zymomonas are reported, but the inhibition of microbial growth by the accumulated acetaldehyde was the main challenge to keep its continuous production. This challenge has been addressed in this study and microbubbles generated by fluidic oscillation were used to remove both acetaldehyde and carbon dioxide from the fermentation broth. Additionally, the oxygenation concurrent with the stripping process by microbubbles efficiently maintained the oxygen concentration in the fermentation broth above the critical oxygen concentration, leading to stable aerobic conditions. The results show that 45 % yield of ethanol and 1 % yield of acetaldehyde with 110 % yield of microbial biomass in comparison with 70 %, 0.5 % and 90 % yield for ethanol, acetaldehyde, and biomass respectively in the initially sparged group were achieved. Also, acetaldehyde was removed from the fermentation broth with 99 % efficiency. Acetaldehyde production in the fermentation was enhanced by selecting the mutant cells with attenuated or modified alcohol dehydrogenase activity using increasing concentrations of allyl alcohol. The results show that 17-fold increase was achieved in the mutant strain in comparison with the wild strain. In addition, the mutant strain produced 90 % less ethanol than the wild VI strain. Also, the acetaldehyde removal efficiency was 88.5 % in comparison with 42 % efficiency achieved with the fine bubbles (bigger bubbles). Additionally, biomass yield produced by the mutant strain was less by a half than the yield produced by the wild strain. To enhance the biomass yield of the mutant strain, different techniques were used to grow this bacterium aerobically, but maintaining sufficient oxygen concentration was challenging in the bacterial propagation stage. Oxygen is the limiting factor in the aerobically grown bacterial cultures, but similarly, the impact of mixing can be critical. The results show that the oxygen uptake rate and mass transfer coefficient are substantial increased using microbubbles technology and there were 41-fold and 150-fold increase in the oxygen uptake rate and mass transfer coefficient respectively in the microbubbles-dosed culture in comparison with the shaking flask culture. This technology can also achieve a proper mixing. Regarding the biomass yield, the mutant strain of Zymomonas mobilis shows an increased yield using the shaking flask (around 100 % and 133 % increases) in comparison with other (microbubbles-dosed and stationary respectively) techniques, while the wild strain produces more biomass in the microbubble-based technique (around 50 % and 100 %) than other (shaking flask and stationary respectively) techniques. In addition, a propagation unit was designed and simulated to grow the mutant strain aerobically in the propagation stage before using this grown biomass as an inoculum to the fermentation process. Fundamentally, the results obtained in this study are achieved in a laminar flow with several orders of magnitude lower energy density than conventional benchmarks, which are a highly turbulent flow.

Proteins are the most abundant biomolecules within a cell and are involved in all biochemical cellular processes ultimately determining cellular function. Therefore, to develop a complete understanding of cellular processes, obtaining knowledge about protein function and interaction at a molecular level is critical. Consequently, the investigation of proteins in their native environment or in partially purified mixtures is a major endeavour in modern life sciences. Due to their high chemical similarity, the inherent problem of studying proteins in complex mixtures is to specifically differentiate one protein of interest from the bulk of other proteins. Site-specific protein functionalisation strategies have become an indispensable tool in biochemical- and cell biology studies. This thesis presents the development of a new enzymatic site-specific protein functionalisation strategy that is based on the reversible covalent phosphocholination of short amino acid sequences in intact proteins. A synthetic strategy has been established that allows access to functionalised CDP-choline derivatives carrying fluorescent reporter groups, affinity tags or bioorthogonal handles. These CDP-choline derivatives serve as co-substrates for the bacterial phosphocholinating enzyme AnkX from Legionella pneumophila, which transfers a phosphocholine moiety to the switch II region of its native target protein Rab1b during infection. We identified the octapeptide sequence TITSSYYR as the minimum recognition sequence required to direct the AnkX catalysed phosphocholination and demonstrated the functionalisation of proteins of interest carrying this recognition tag at the N- or C-terminus as well as in internal loop regions. Moreover, this covalent modification can be hydrolytically reversed by the action of the Legionella enzyme Lem3, which makes the labeling strategy the first example of a covalent and reversible approach that is fully orthogonal to current existing methodologies. Thus, the here presented protein functionalisation approach holds the potential to increase the scope of possible labeling strategies in complex biological systems. In addition to the labeling of tagged target proteins, a CDP-choline derivative equipped with a biotin affinity-tag was synthesised and used in pull-down experiments to investigate the substrate scope of AnkX and to elucidate the role of protein phosphocholination during Legionella pneumophila infection.
Proteiner utgör huvudbeståndsdelen av alla biomolekyler i en cell. Dessa är involverade i alla cellulära processer som bestämmer cellens egenskaper. För att förstå de cellulära processerna är det nödvändigt att förstå proteinernas funktion på molekylär nivå. Att studera proteiner i deras naturliga omgivning, det vill säga inuti en cell eller i ett cellextrakt, är en stor utmaning i dagens livsvetenskaper. Eftersom proteiner är kemiskt lika varandra så är det svårt att skilja ett från tusentals andra. Att specifikt märka proteiner för att skilja ut dem från bakgrunden har blivit ett viktigt arbetssätt i modern biokemi och cellbiologi. Avhandlingen beskriver utvecklandet av en ny metod för reversibel och kovalent enzymatisk märkning baserat på fosfokolinering/defosfokolinering av en kort aminosyrasekvens i intakta proteiner. En syntesmetod för att framställa onaturliga CDP-kolinderivat har etablerats vilket tillåter oss att framställa CDP-kolin som bär en funktionalitet, vilket kan vara ett färgämne eller en affinitetstagg. Dessa onaturliga CDP-kolinderivat accepteras som co-substrat av enzymet AnkX från Legionella pneumophila vilket transfererar den funktionaliserade delen av CDP-kolinderivatet till en kort aminosyrasekvens baserad på AnkX’s naturliga substrat vid infektion, det lilla GTPaset Rab1. Under avhandlingsarbetets gång identifierades den kortaste aminosyrasekvensen som känns igen av AnkX, endast de åtta aminosyrorna TITSSYYR är nödvändiga för igenkänning av AnkX. Dessa åtta aminosyror kan genetiskt infogas i början, slutet eller mitt i ett protein för igenkänning och funktionalisering via AnkX och våra syntetiska CDP-kolinderivat. Vid Legionellainfektion i eukaryota celler klyvs fosfokolineringen efter en viss tid, eftersom Legionella pneumophila producerar ett fosfodiesteras, Lem3, som tar bort de fosfokolineringar som AnkX har installerat när de inte längre behövs. Vi har använt Lem3 för att ta bort märkning i sekvensen TITSS(PC)YYR, vilket gör vår strategi helt reversibel. Vi har kunnat demonstrera att AnkX-Lem3 systemet accepterar ett brett spektrum av CDP-kolinderivat, vilket gör metoden till den första av sitt slag, eftersom den är fullt reversibel. Vi har vidare undersökt vilka proteiner AnkX reagerar med inuti celler, vi använde oss av ett CDP-kolinderivat funktionaliserat med biotin, vilket har tillåtit oss att fiska ut alla de proteiner som fosfokolineras av AnkX. Förutom de små GTPaserna i Rab-familjen så identifierade vi även IMPDH2, ett enzym som reglerar det hastighetsbestämmande steget i syntesen av guanosin-nukleotider. Detta är mycket intressant, eftersom det leder till frågan ifall Legionella pneumophila manipulerar sin värdcell genom att förändra mängden GTP i förhållande till ATP.

Thesis (M.S. in Management)--Naval Postgraduate School, June 2007.
Thesis Advisor(s): William R. Gates, Peter J. Coughlan. "June 2007." Includes bibliographical references (p. 67-68). Also available in print.

Outcome dependent sampling may increase efficiency in observational studies. It is however not always obvious how to sample efficiently, and how to analyze the resulting data without introducing bias. This thesis describes a general framework for efficiency calculations in multistage sampling, with focus on what is sometimes referred to as ascertainment sampling. A method for correcting for the sampling scheme in analysis of ascertainment samples is also presented. Simulation based methods are used to overcome computational issues in both efficiency calculations and analysis of data.
At the time of doctoral defense, the following paper was unpublished and had a status as follows: Paper 1: Submitted.

In the human visual system, one of the most prominent functions of the extensive feedback from the higher brain areas within and outside of the visual cortex is attentional modulation. The feedback helps the brain to concentrate its resources on visual features that are relevant for recognition, i. e. it iteratively selects certain aspects of the visual scene for refined processing by the lower areas until the inference process in the higher areas converges to a single hypothesis about this scene. In order to minimize a number of required selection-refinement iterations, one has to find a short sequence of maximally informative portions of the visual input. Since the feedback is not static, the selection process is adapted to a scene that should be recognized. To find a scene-specific subset of informative features, the adaptive selection process on every iteration utilizes results of previous processing in order to reduce the remaining uncertainty about the visual scene. This phenomenon inspired us to develop a computational algorithm solving a visual classification task that would incorporate such principle, adaptive feature selection. It is especially interesting because usually feature selection methods are not adaptive as they define a unique set of informative features for a task and use them for classifying all objects. However, an adaptive algorithm selects features that are the most informative for the particular input. Thus, the selection process should be driven by statistics of the environment concerning the current task and the object to be classified. Applied to a classification task, our adaptive feature selection algorithm favors features that maximally reduce the current class uncertainty, which is iteratively updated with values of the previously selected features that are observed on the testing sample. In information-theoretical terms, the selection criterion is the mutual information of a class variable and a feature-candidate conditioned on the already selected features, which take values observed on the current testing sample. Then, the main question investigated in this thesis is whether the proposed adaptive way of selecting features is advantageous over the conventional feature selection and in which situations. Further, we studied whether the proposed adaptive information-theoretical selection scheme, which is a computationally complex algorithm, is utilized by humans while they perform a visual classification task. For this, we constructed a psychophysical experiment where people had to select image parts that as they think are relevant for classification of these images. We present the analysis of behavioral data where we investigate whether human strategies of task-dependent selective attention can be explained by a simple ranker based on the mutual information, a more complex feature selection algorithm based on the conventional static mutual information and the proposed here adaptive feature selector that mimics a mechanism of the iterative hypothesis refinement. Hereby, the main contribution of this work is the adaptive feature selection criterion based on the conditional mutual information. Also it is shown that such adaptive selection strategy is indeed used by people while performing visual classification.

Depuis longtemps les réactifs isothiocyanates (ITCs) sont largement utilisés dans le domaine de la bioconjugaison. Ces électrophiles forts réagissent avec les cystéines et les lysines des protéines pour former une liaison stable. Cette réactivité click permet de réaliser des marquages sélectifs ainsi que des fonctionnalisations de protéines. Cependant, les ITCs ne sont pas faciles à synthétiser et à isoler et leur stabilité ne permet pas une conservation optimale.Le but de ce projet est de développer le système enzymatique myrosinase-glucosinolate (MG) comme outil de conjugaison capable de former un ITC in situ. Le tandem MG est un mécanisme de défense des plantes de l’ordre des Brassicales bien connu. Dans ce système biochimique, la myrosinase opère comme thioglucosidase, en hydrolysant les glucosinolates (GLs), pour générer des ITCs. L’avantage de ce tandem enzymatique est de produire les ITCs à partir de précurseurs solubles dans l’eau, non toxiques, sous conditions douces.Afin d’explorer cet outil enzymatique, deux types de GLs non naturels ont été conçus. En raison de l’importance de l’interaction lectine-mannose dans les mécanismes d’adhésion bactérienne, nous avons conçu une petite librairie de GLs intégrant un mannoside. De cette façon il est possible d’étudier des lectines bactérienne (FimH). Le second type de glucosinolate est caractérisé par une deuxième fonction chimique, permettant de réaliser des réactions orthogonales.Le système MG a été évalué dans plusieurs approches de bioconjugaison telles que le marquage sélectif d’une lectine, la synthèse de néoglycoprotéines et la fonctionnalisation de nanoparticules
Since many decades, Isothiocyanate (ITCs) reagents are widely used in bioconjugation approaches to create a stable bond onto the lysin and cysteine residues of proteins and peptides. Thanks to this click reaction it is possible to achieve a selective ligation or a high functionalization of proteins. On the others hand, isothiocyanates are not easy to prepare, to isolate, to store and most of the time insoluble in water.The aim of this work is to explore the myrosinase-glucosinolate (MG) enzymatic tandem as a ligation system, in order to release ITCs in-situ. The MG tandem is a well-known mechanism of defense in plants of the order Brassicales. In this biochemical system, myrosinase acts as a thioglucosidase, hydrolyzing glucosinolates (GLs) to liberate transient species that spontaneously form ITCs. The advantage of this enzymatic tandem is to generate ITCs from a stable non-toxic GLs precursor, soluble in water, using mild conditions.In order to develop this enzymatic tool, two kind of unnatural GLs were synthetized. Due to the important role of the mannoside-lectins interaction in the bacterial adhesion mechanism, at first, we designed a small library of GL bearing mannosides, in order to target and study the bacterial lectin FimH. The second kind of GL possess a chemical function allowing to study orthogonal reactions. After, the ability of the myrosinase to hydrolyze those unnatural GLs was investigated, as well their chemical reactivity.Then, the performances of the MG system was studied in different approaches of bioconjugation, such as: the synthesis of neoglycoproteins (NGPs), the site selective labeling of a lectin and the functionalization of gold nanoparticles. The feasibility of these strategies confirmed that the MG system can be used as an enzymatic tool in some bioconjugation approaches

Antigene e anticorpo sono i due reagenti chiave di un saggio immunodiagnostico. L’indagine di nuove tecniche e il miglioramento di processi quali la purificazione e la marcatura sito-specifica di antigeni e anticorpi possono promuovere lo sviluppo di nuovi reagenti più efficienti capaci di migliorare la performance dei saggi immunodiagnostici. La prima parte di questa tesi è stata focalizzata sull’esplorazione di tecniche biotecnologiche innovative nella produzione di antigeni al fine di migliorare i saggi per la rilevazione degli anticorpi IgM e IgG specifici per il virus Epstein-Barr. Il virus EBV è causa della mononucleosi infettiva ed è associato ad un crescente numero di tumori; per questa motivo è importante sviluppare saggi diagnostici per la rilevazione di EBV ad alta specificità e sensibilità. La proteina VCA p18 è uno degli antigeni più importanti per la diagnosi di EBV. I saggi attuali Diasorin LIAISON EBV VCA IgM and IgG si basano su un singolo antigene corrispondente alla regione C-terminale immunodominante della proteina p18, immobilizzata su fase solida. I vari metodi esplorati in questa tesi hanno permesso di ottenere diverse varianti dell’antigene p18 con lo scopo di migliorare le prestazioni dei saggi EBV VCA IgM e IgG a diversi livelli: 1_produzione dell’antigene p18; 2_immobilizzazione dell’antigene p18 su fase solida; 3_formato di saggio. 1_La lunghezza della regione C-terminale della proteina p18 (57aa), risulta essere considerevole per il processo sintetico ma, allo stesso tempo, troppo piccola per essere prodotta in modo efficiente per via ricombinante. Per superare questo problema, abbiamo esplorato il sistema Elastin Like Polypeptides (ELP)-Inteina basato sull’uso di una proteina capace di effettuare auto-cleavage (inteina) e un tag responsivo alla temperatura (ELP). Questa tecnica si è rivelata un eccellente sistema per la produzione del peptide p18. 2_Lo sviluppo di diverse varianti dell’antigene p18 ha permesso di esplorare varie tecniche per l’immobilizzazione dello stesso antigene su fase solida: coating covalente diretto, attraverso il complesso streptavidina-biotina e attraverso l’uso dei peptidi leucine zipper (o velcro). L’immobilizzazione del peptide p18 su fase solida attraverso questi vari metodi è avvenuta con successo e l’attività immunochimica dell’antigene, immobilizzato con queste tecniche innovative, è comparabile o migliore rispetto a quella del peptide sintetico utilizzato nei saggi attuali. 3_Nonostante il saggio Diasorin LIAISON EBV VCA IgM abbia una buona performance analitica, al fine di ottenere un aumento di specificità, è stato esplorato un nuovo tipo di formato di saggio. Sfortunatamente i risultati indicano che questo diverso tipo di formato raggiunge un livello di specificità minore rispetto a quello del saggio attuale. La seconda parte di questa tesi è stata focalizzata sull’esplorazione di un metodo innovativo per la marcatura sito specifica degli anticorpi. Uno degli approcci più promettenti è basato sulla generazione di gruppi tiolo liberi attraverso la riduzione parziale e selettiva dei ponti disolfuro inter-catena presenti a livello della “hinge region” e la loro reazione con molecole marcanti caratterizzate dal possedere gruppi funzionali reattivi verso i gruppi sulfidrilici. Questa tecnologia è stata utilizzata per la biotinilazione di due diversi anticorpi usati attualmente per il rilevamento della proteina virale p24 di HIV e per l’antigene FGF23. I risultati suggeriscono che la biotinilazione sito-specifica rispetto a quella classica random promuove un miglioramento dell'attività immunochimica degli anticorpi con una conseguente ottimizzazione della performance dei saggi immunodiagnostici.
Antigen and antibody are the two key reagents for an immunodiagnostic assay. Investigation of new techniques and improvement of processes such as purification and site-specific labeling of antigen and antibody molecules can promote the development of new more powerful bioreagents able of improving the performance of immunodiagnostic assays. The first part of this thesis aimed to explore innovative biotechnology techniques in antigen production for the improvement of immunoassays that allow the detection of antibodies directed against the Epstein-Barr virus. EBV virus is the causative agent of infectious mononucleosis and it is considered to be associated with a still increasing number of tumors; for this reason it is important to develop diagnostic assays for EBV detection with high specificity and sensitivity. The viral capsid protein VCA p18 is one of the most important antigens for the diagnosis of EBV. The current Diasorin LIAISON EBV VCA IgM and IgG assays rely on a single antigen, consisting in a synthetic peptide corresponding to the immunodominant C-terminal portion of the p18 protein, which is immobilized on solid phase (indirect format). The several methods explored in this thesis have allowed to obtain different variants of the p18 antigen with the aim to improve the performance of DiaSorin LIAISON EBV VCA IgM and IgG assays at different levels: 1_production of p18 antigen; 2_immobilization of p18 antigen on solid phase; 3_immunoassay format. 1_ The length of the immunodominant C-terminal portion of the p18 protein (57aa) appears to be considerable for the synthetic route but, at the same time, too small to be effectively produced in a recombinant fashion. To overcome this problem, we explored the Elastin Like Polypeptides (ELP)-Intein system, a method based on the use of a self-cleavable protein (the intein) and a temperature responsive tag (ELP). This technique has proved to be an excellent system for the preparation of the p18 peptide. 2_The development of different variants of p18 antigen has enabled to explore different techniques for the immobilization of the same antigen on solid phase: direct covalent coating, through streptavidin-biotin complex and through an innovative procedure based on the use of leucine zipper (or “velcro”) peptides. The immobilization of the p18 peptide on solid phase through these different methods occurred successfully and the immunochemical activity of the antigen, immobilized with these innovative techniques, is comparable or better than that of the synthetic peptide used in the current immunoassays. 3_Despite the DiaSorin LIAISON EBV VCA IgM immonoassay has a good analytical performance, in order to obtain an increase of specificity, a new assay format was explored. Unfortunately, the results indicate that this different type of format reaches a lower level of specificity than that of current assay. The second part of this thesis aimed to explore an innovative method for the site-specific labeling of antibodies. One the most promising approach is based on the generation of free thiol groups by selective partial reduction of the interchain disulfide bridges present at the level of the “hinge region” and their reaction to labels carrying sulfhydryl-reactive chemical groups. This technology was used for the biotinylation of two different antibodies currently used in immunoassays for the HIV p24 viral protein and for FGF23 antigen detection. The results suggest that the site-specific biotinylation compared to the random traditional biotinylation promotes a great improvement of antibodies immunochemical activity with a consequent optimization of immunodiagnostic assays performance.

Character recognition is not a difficult task for humans who repeat the process thousands of times every day as they read papers or books. However, after more than 40 years of intensive investigation, there is still no machine that can recognize alphabetic characters as well as humans. Optical Character Recognition (OCR) is the process of converting a raster image representation of a document into a format that a computer can process. It involves many sub-disciplines of computer science including digital image processing, pattern recognition, natural language processing, artificial intelligence, and database systems. Applications of OCR systems are broad and include postal code recognition in postal departments, automatic document entries in companies and government departments, cheque sorting in banks, machine translation, etc. The objective of this thesis is to design an optical character recognition system which can recognize Arabic script. This system has to be: 1) accurate: with a recognition accuracy of 953; 2) robust: able to recognize two different Arabic fonts; and 3) efficient: it should be a real time system. This proposed system is composed of five image processing processes: 1) Image Acquisition; 2) Preprocessing; 3) Segmentation; 4) Feature Extraction; and 5) Classification. The recognized results are presented to users via a window-based user-interface. Thus, they can control the system, recognize and edit documents by a click on the mouse button. A thinning algorithm, a word segmentation algorithm and a recognition based character segmentation algorithm for Arabic script have been proposed to increase the recognition accuracy of the system. The Arabic word segmentation algorithm successfully segments the horizontally overlapped Arabic words, whereas the recognition-based character segmentation algorithm replaces the classical character segmentation method and raises the level of accuracy of recognition of the proposed system. These blocks have been integrated. Results to test the requirements of accuracy, robustness and efficiency are presented. Finally, some extensions to the system have also been proposed.

The research activities of my doctorate have been devoted to the application of in-cell NMR for the characterization of proteins directly inside human cells. Within this context, I have also addressed the development of a methodology to perform sequential labeling for in-cell NMR in human cells. The in-cell NMR approach, which consists in observing one or more selectively labeled protein(s) or nucleic acid in living cells through high resolution NMR experiments, is a powerful tool to obtain structural and functional information in situ, thus overcoming the limitation of studying macromolecules isolated from the other cellular components. Protein folding, posttranslational modifications and interactions with specific partners or cellular elements can be studied in the physiological environment in which the protein of interest operates. My research work also addressed the in vitro characterization of the phosphorylation pattern of the non-structural protein 5A domain 3 of Hepatitis C virus by casein kinase 2.

碩士
國立政治大學
資訊科學學系
104
Difangzhi is the local gazetteers compiled by local government of China. Its content is plenty and extensive. It’s including many undetected information, like biographical information, geographical information, and officer record information and so on. Because of the difference between Difangzhi corpus and modern Chinese language, we should not use current natural language processing tools directly. In order to extract biographical information, we construct our model to recognize the named entity and use the noun list to assist our annotation method in Difangzhi corpus. In this study, we use supervised learning to construct our model. At first, we need to generate our training data. According to the personal information list with manual verification and noun lists, we have reliable information to annotate words in Difangzhi corpus. However, they still have some noise in those lists. As a result, we must do the preprocessing to those lists for cleaning. After, the ambiguity problem will happen when we trying to annotate our corpus. Here we provide three methods to annotate our corpus with disambiguation. Using the annotated corpus to generate training data and built the condition random fields models. In our experiment, we use our models generated by three different annotate methods to predict the character label in testing Difangzhi corpus. According to the labeled result, we extract the person name and address name to evaluate. The result shows the precision of person name recognition is over 80%, and precision of address name recognition is about 86%. Because of the training corpus and test corpus is quite similar, the performances of our model is pretty well. Therefore, we use labeled result to find correlation of person name and address name. Using a simple way to connect person name and address name and sampling the result to evaluate. The sample result shows we could connect person name and address name correctly in some specific grammars. In order to analyze more deeply, we attempt to split clauses in Difangzhi corpus. Use finite state machine model to recognize the beginning of clauses. Although the result shows we could find some beginning of clauses, but our method still lose many beginning of clauses. In the future work, we attempt to add more information to annotate Difangzhi corpus and modify our disambiguated methods to make the recognition result better. In order to get more information about the person in the corpus, we will try to split paragraphs or sentences more precisely. Besides, we also try to analyze grammar in the corpus. Finding useful pattern to connect person name and other entities, like address name, officer name and so on. Generating the information about people appears in the corpus automatically.

碩士
國立中正大學
化學暨生物化學研究所
100
A steroselectivity synthesis of fully subsititude decahydro-1H- pentaleno [2,1-c]pyran derivatives have been achieved with alkylidene malonates and aldehydes. By a sequential oganocatalyzed anti-selective Michael reaction－reduction－lactonization－Pauson-Khand reaction, varios additives effected different results of steroselectivity.

Penetrating keratoplasty is a full-thickness corneal transplant procedure with a relatively long post-operative visual rehabilitation period. Post-operative corneal astigmatism often causes refractive error resulting in suboptimal vision, despite a clear graft. In order to reduce this issue, surgeons selectively remove sutures from the transplanted cornea to manipulate and control the levels of post-operative astigmatism present. In order to identify tight sutures causing astigmatism, corneal topography instruments have been developed which use reflected light patterns to reconstruct topographical images of the corneal surface and provide measurements of corneal steepness and astigmatism. Currently, standard conventions limit suture removal to one to two sutures per visit at an interval between four and six weeks. This experiment sought to determine the feasibility of multiple suture removal following corneal transplant by evaluating the change in astigmatism occurring immediately after suture removal in corneal transplant patients and comparing the change to any occurring one month later. Four separate samples were obtained and analyzed to determine if topography-based decision immediately post-suture removal matched suture removal decision one month later. Topography-guided decisions immediately following suture removal incorrectly identified the appropriate subsequent suture in all samples. Data was analyzed using Fisher’s exact test to determine statistical probability of results, and there was a statistically significant difference between topography-based decision immediately after suture removal and standard topography-based decision at one month. This demonstrated that the period of time immediately following suture removal was not reliable in determining the correct subsequent suture to be removed. Limitations of this study included a small patient sample size, potential graft-host junction override in samples, and the weight of subjective determination by the surgeon. While penetrating keratoplasty has been found to be an effective treatment for patients, further research is warranted to investigate the timeline behind corneal astigmatic stability following surgery and to identify opportunities to shorten long rehabilitation periods.

Carbohydrates and proteins represent two large groups of biomolecules which are tremendously important for biological processes in health and disease state. Although protein-structures are encoded in the genome, cellular glycan structures are template independent and can only be addressed in an indirect manner. The development of metabolic oligosaccharide engineering (MOE) gave rise to new methods to study carbohydrate structures in the context of different disease settings and in different organisms. While in many cases mannose derivatives are used to study the sialic acid structures in cancer cells, this work presents the results on the metabolic incorporation of galactose derivatives into cell membrane glycans of human hepatic cells. Three unnatural galactose derivatives containing terminal alkene groups in C2 or C6 position were synthesized and their reaction rates in inverse electron demand Diels Alder reactions (iEDDA) were evaluated, by using a high-throughput screening method in 96-well plates. It was shown that none of the developed galactose derivatives exhibit any cell toxic effect in HepG2 or Huh7 cell lines. Furthermore, all monosaccharides could be successfully incorporated in cell membrane glycan structures of both cell lines and the localization on the cell membrane was confirmed by co-localization with a plasma membrane dye. After developing this incorporation and labeling strategy of unnatural galactose derivatives in the cell membrane of human hepatic cells, the change in incorporation during an infection of these cells with Plasmodium berghei sporozoites was investigated. By using different techniques, such as confocal microscopy, flow cytometry and imaging flow cytometry, only a small trend for an increased uptake of the unnatural galactose derivative in P. berghei infected cells was observed. To explain this result, the pathway for the diffusion of the unnatural galactose derivative was determined. The application of specific and non-specific inhibitors for the glucose transporter GLUT1 revealed that this transporter is involved in the delivery of galactose derivatives into cultured cells. The enhanced translocation of this transporter to the surface of infected hepatic cells explains the observed tendency for an increased incorporation of the unnatural galactose derivative in these cells. Apart from cell studies, MOE was applied for the first time to study a possible transfer of galactose monosaccharides from the mosquito host to the parasite. Biosynthetic pathways for glycan assembly in the parasite are poorly understood. Suggestions on the participation of the mosquito host in some of these pathways, led to the idea to apply MOE in this situation. It was possible to show an uptake of the presented galactose derivatives by the mosquito but only reduced transfer to the parasite seems to occur. In addition to the development of monofunctional galactose derivatives, also a bifunctional derivative containing two orthogonal reporter groups was synthesized. However, so far it was not possible to achieve a metabolic incorporation or labeling of this derivative on cell membrane glycans. After developing cellular tools to study carbohydrate structures, a site-selective method for protein modification was generated, to be used for the development of new glycoconjugate vaccine candidates. By introducing selectively two dehydroalanine residues in place of the disulfide bond C186-C201 of the immunogenic protein CRM197, a new chemical moiety for the conjugation of carbohydrate antigens was obtained. It was shown that these moieties can be used for the selective introduction of polysaccharide antigens from group B Streptococcus (GBS) or Streptococcus pneumoniae. Both types of glycoconjugates could be synthesized and first trials on the purification methods were undertaken. This concept will be developed further for future vaccine candidates. Finally, a synthetic method was developed which could facilitate the synthesis of defined antigenic oligosaccharide structures. This method uses the thiophilic promoter O-mesitylenesulfonylhydroxylamine (MSH) for the activation of thioglycoside donors. It was demonstrated that different thioglycoside donors are activated with different kinetics, depending on the presented protecting groups or the anomeric leaving group. Apart from applying the developed activation method for the synthesis of several glycosylation products, the sequential activation of S-alkyl before S-phenyl anomeric groups was shown during the synthesis of a model trisaccharide. Overall, bio-orthogonal methods were developed and applied for the investigation of carbohydrate structures in the context of malaria disease, and for the site-selective modification of protein carriers during the development of glycoconjugate vaccine candidates.

碩士
國立中正大學
化學所
94
Radioactive cesium contamination of water is of serious social and environmental concern since it is a significant fraction of the radioactivity of the liquid waste. Nickel hexacyanoferrate (NiHCF) has been widely studied for separation and electrochemical sensing of Cs+ owing to their strong selectivity for Cs+ over other alkali cations. In this work, a novel material, multilayers of NiHCF immobilized within or on a porous propyl-ethylene diamine triacetate functionalized silica matrix, has been synthesized and characterized. Its uses as a novel cesium sorbent and cesium sensing material have been studied. As a cesium sorbent, this material was evaluated for selective removal of cesium ions in the presence of competing metal ions from water. Loading capacity of at least 1.69 mmole per g of sorbent material has been achieved. As a cesium sensing material, multilayers of NiHCF were coated on glassy carbon electrodes by multiple sequential adsorption for potentiometric sensing of cesium ions. A Nernst response with a slope of 31 mV/decade was obtained between 10-6 M and 1 M. The material was characterized through infrared spectroscopy, X-ray powder diffraction, cyclic voltammetry, X-ray energy dispersive spectroscopy and the Ni and Fe precursors were modeled by the PM3TM semiempirical method.

博士
國立成功大學
基礎醫學研究所
94
Integrins are alpha/beta�nheterodimers on cellular membrane and involved in cell adhesion, mobility, and signal transduction. Many physiological and non-physiological integrin ligands utilize RGD, LDV, or a related sequence as a key structural component of their integrin-binding site. Rhodostomin (Rho) is a potent integrin antagonist, which consists of 68 amino acids including six disulfide bonds and a PRGDMP sequence at the positions of 48-53. Rho competes against fibrinogen in the binding of integrin �娗lphaIIbbeta3 on platelet and results in inhibiting platelet aggregation. Many studies have also demonstrated that the amino acid residues flanking the RGD sequence of integrin ligands modulate their specificity of interaction with integrin alphaIIbbeta3,alphaVbeta3, and alpha5beta1 complexes. In order to identify the amino acid residues required for selective recognition of integrins alphaIIbbeta3, alphaV�涀eta3, and�nalpha5beta1, we mutated the residues in the RGD loop or C-terminal region of Rho, expressed Rho mutants in P. pastoris, and used the platelet aggregation and cell adhesion assays for functional studies. We found that Rho mutants containing the sequences AKGDWN, as well as RLD and RGDD can selectively inhibit integrins alphaIIbbeta3 and alphaVbeta3, respectively. In addition, the mutants containing the sequences ARGDGW and ARGDXX have better selectivity or activity in inhibiting integrin alpha5beta1. 3D structures of integrin-specific Rho mutants were determined by NMR spectroscopy. Our results showed that the orientations of R49 and D51 residues and dynamic properties of the RGD motif in Rho play important roles in integrin recognition. To develop the protocol to produce amino-acid-type selective (AATS) isotope labeling of protein expressed in P. pastoris, Rho was used to optimize the culture condition. The optimized protocol was successfully applied to produce AATS isotope-labeled Rho and other protein, dendroaspin. The labeling of [alpha-15N]-Cys, -Leu, -Lys, and -Met amino acids has an incorporation rate greater than 50%. In addition, the culture condition was successfully applied to label dendroaspin, a four disulfide-bonded protein expressed in P. pastoris. This is the first study to present a protocol for AATS isotope labeling of protein expressed in P. pastoris. This study will serve as the basis for exploring structure-activity relationships of the integrin/disintegrin complexes as well as rational drug design on integrins.

The cytoskeletal protein vinculin plays a key role in the control of cell-cell or cell-matrix adhesions. It is involved in the assembly and disassembly of focal adhesions and affects their mechanical stability. While many facts highlight the importance and significance of vinculin for vital processes, its precise role in the regulation of cell adhesions is still only partially understood. Various EPR methods are used in this work in order to study the vinculin tail (Vt) domain in an aqueous buffer solution and its structural changes induced by F-actin and acidic phospholipids. EPR results in combination with a rotamer library approach (RLA), MD simulation and other computational methods allowed the construction of molecular models of Vt and dimeric Vt in the presence and absence of its binding partners. Furthermore, X-band orientation selective DEER measurements were applied on a Vt double mutant. It could be shown that the determination of the mutual orientation of protein bound spin labels is possible at X-band frequencies, if the orientation correlation of the spin label pair is strong. The method established here can be used to determine valuable information about proteins and nucleic acids, expanding the virtue of DEER spectroscopy as a tool for structure determination.

In the last decade, there has been an explosive growth in different applications of wireless technology, due to users' increasing expectations for multi-media services. With the current trend, the present systems will not be able to handle the required data traffic. Lattice codes have attracted considerable attention in recent years, because they provide high data rate constellations. In this thesis, the applications of implementing lattice codes in different communication systems are investigated. The thesis is divided into two major parts. Focus of the first part is on constellation shaping and the problem of lattice labeling. The second part is devoted to the lattice decoding problem. In constellation shaping technique, conventional constellations are replaced by lattice codes that satisfy some geometrical properties. However, a simple algorithm, called lattice labeling, is required to map the input data to the lattice code points. In the first part of this thesis, the application of lattice codes for constellation shaping in Orthogonal Frequency Division Multiplexing (OFDM) and Multi-Input Multi-Output (MIMO) broadcast systems are considered. In an OFDM system a lattice code with low Peak to Average Power Ratio (PAPR) is desired. Here, a new lattice code with considerable PAPR reduction for OFDM systems is proposed. Due to the recursive structure of this lattice code, a simple lattice labeling method based on Smith normal decomposition of an integer matrix is obtained. A selective mapping method in conjunction with the proposed lattice code is also presented to further reduce the PAPR. MIMO broadcast systems are also considered in the thesis. In a multiple antenna broadcast system, the lattice labeling algorithm should be such that different users can decode their data independently. Moreover, the implemented lattice code should result in a low average transmit energy. Here, a selective mapping technique provides such a lattice code. Lattice decoding is the focus of the second part of the thesis, which concerns the operation of finding the closest point of the lattice code to any point in N-dimensional real space. In digital communication applications, this problem is known as the integer least-square problem, which can be seen in many areas, e.g. the detection of symbols transmitted over the multiple antenna wireless channel, the multiuser detection problem in Code Division Multiple Access (CDMA) systems, and the simultaneous detection of multiple users in a Digital Subscriber Line (DSL) system affected by crosstalk. Here, an efficient lattice decoding algorithm based on using Semi-Definite Programming (SDP) is introduced. The proposed algorithm is capable of handling any form of lattice constellation for an arbitrary labeling of points. In the proposed methods, the distance minimization problem is expressed in terms of a binary quadratic minimization problem, which is solved by introducing several matrix and vector lifting SDP relaxation models. The new SDP models provide a wealth of trade-off between the complexity and the performance of the decoding problem.